{
  "version": 3,
  "sources": ["ssg:https://framerusercontent.com/modules/RJN7voly8d65PgTrzGHQ/XKexiIJk6O4YtXRuM9GT/VrN4RwcK9.js"],
  "sourcesContent": ["import{jsx as e,jsxs as i}from\"react/jsx-runtime\";import{addPropertyControls as t,ComponentPresetsConsumer as n,ControlType as a,Link as r}from\"framer\";import{motion as o}from\"framer-motion\";import*as s from\"react\";import{Youtube as l}from\"https://framerusercontent.com/modules/NEd4VmDdsxM3StIUbddO/bZxrMUxBPAhoXlARkK9C/YouTube.js\";let c=\"gYtMT1cTw\",h=\"A8BMQitKd\",d=\"J2ZD_SFeL\",m=\"wr9uZn_Dd\",p=\"oVU11fX32\",g=\"ueAVa7IXk\",u=\"M_CYyp3j6\",f=\"WSrp1qLr8\",w=\"L5oJzTTDV\",y=\"q6EWbj318\",b=\"kXwi28nMo\",v=\"vaQJHLM83\",I=(e,i)=>{if(e&&\"object\"==typeof e)return{...e,alt:i};},A=[{id:\"IqTm5zFcG\",[c]:\"How AI is Bolstering Chinese Espionage Programs\",[h]:\"how-ai-is-bolstering-chinese-espionage-programs\",[d]:\"2024-01-02T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/6AyW8SmfAq71Ygaj43dc2GyDps.png\",srcSet:\"https://framerusercontent.com/images/6AyW8SmfAq71Ygaj43dc2GyDps.png?scale-down-to=512 512w, https://framerusercontent.com/images/6AyW8SmfAq71Ygaj43dc2GyDps.png 1024w\"},\"\"),[p]:\"How AI is Bolstering Chinese Espionage Programs\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"In July 2018, Xiaolang Zhang, a former Apple employee, was arrested at San Jose airport en route to Beijing. He was charged with stealing trade secrets from Apple's autonomous-driving program. The case of Zhang, who pleaded guilty to stealing trade secrets from Apple, is just a small part of a much larger picture. U.S. authorities suspect that Chinese intelligence is using stolen data from various sources, including the Office of Personnel Management and health insurers, to track U.S. spies and officials. Recent warnings from U.S. intelligence suggest that China might use AI not just for intellectual property theft but also for large-scale data collection on Americans. F.B.I. Director Christopher Wray has expressed concerns about AI amplifying China's hacking operations, using American technology against the U.S.\"}),/*#__PURE__*/e(\"p\",{children:\"Over the past decade, Beijing has been linked to massive data hacks, including those at Marriott International, Equifax, and Anthem, among others. These thefts, involving personal information of millions of Americans, have raised alarms about China's capability to use AI to analyze and exploit these vast data sets.\"}),/*#__PURE__*/e(\"p\",{children:'The situation is complex, with China denying involvement in hacking U.S. networks. The Chinese Foreign Ministry has countered U.S. accusations by labeling the U.S. as the \"biggest hacking empire.\" Chinese intelligence officials themselves even admire the U.S. for its digital espionage programs, openly discussing which tactics they could mimic and implement to improve their own programs.'}),/*#__PURE__*/e(\"p\",{children:\"Recently, The Ministry of State Security (M.S.S.), China\u2019s principal intelligence agency, which has both domestic and foreign intelligence responsibilities (think of it as the F.B.I. and C.I.A. combined), has been using AI to develop comprehensive profiles on individuals of interest in Beijing's embassy district. They analyze behavior patterns and vulnerabilities by integrating data from various sources, including surveillance cameras, car license plates, and cellphone data.\"}),/*#__PURE__*/e(\"p\",{children:\"This shift towards advanced technology marks a significant transformation for the M.S.S., which historically relied on more traditional methods of information gathering. The agency, once known for gathering intel through social interactions at embassy events, is now at the forefront of technology-driven espionage, rivaling the C.I.A. in the U.S.\"}),/*#__PURE__*/e(\"p\",{children:\"The U.S., recognizing the strategic importance of these technologies, has doubled its spending on intelligence activities in China under the Biden administration. The C.I.A. has shifted its focus to gathering information on Chinese companies involved in developing cutting-edge technologies, a change from its previous focus on more traditional defense intelligence.\"}),/*#__PURE__*/e(\"p\",{children:\"Despite the technological advancements, the M.S.S. maintains a comprehensive approach to intelligence, combining foreign intelligence gathering with domestic surveillance and political control. Its role in defending the Chinese Communist Party against perceived threats is a top priority. The agency has adapted to the digital age, utilizing social media and other platforms to disseminate its messages and reinforce its objectives.\"}),/*#__PURE__*/e(\"p\",{children:\"The rivalry between the U.S. and China's intelligence agencies echoes the Cold War-era competition between the C.I.A. and the K.G.B. However, the current context is distinct due to China's economic growth and its ability to harness emerging technologies like AI, which the Soviet Union could not match during the Cold War.\"}),/*#__PURE__*/e(\"p\",{children:\"This technological arms race has significant implications for global security and espionage. As AI and other advanced technologies become more integral to intelligence operations, the nature of espionage is transforming, leading to new challenges and requiring innovative approaches from intelligence agencies worldwide.\"})]}),[u]:\"\",[f]:!0,[w]:\"CTvKjSNWG\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"RJBAzvD79\",[c]:\"Huawei's Rise in the Semiconductor Industry\",[h]:\"huawei-s-rise-in-the-semiconductor-industry\",[d]:\"2023-12-18T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/SbeY4noScPlHaYpMi8BgeDmE.png\",srcSet:\"https://framerusercontent.com/images/SbeY4noScPlHaYpMi8BgeDmE.png?scale-down-to=512 512w, https://framerusercontent.com/images/SbeY4noScPlHaYpMi8BgeDmE.png 1024w\"},\"\"),[p]:\"Huawei's Rise in the Semiconductor Industry\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(r,{href:\"https://www.ft.com/content/327414d2-fe13-438e-9767-333cdb94c7e1\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:/*#__PURE__*/e(\"em\",{children:\"The Financial Times\"})})}),\" reported on the moves Huawei has made in recent years, backed by the Chinese government, to grow its semiconductor development capabilities to compete with the US in order to reduce reliance on other countries to produce powerful AI chips.\"]}),/*#__PURE__*/e(\"p\",{children:\"In late 2020, Huawei, a major player in the mobile phone industry, faced significant challenges due to sanctions imposed by the Trump administration. These sanctions severed Huawei's access to global semiconductor supply chains, hindering its ability to procure chips for advanced handsets. To navigate these challenges, Huawei formed a strategic partnership with the Semiconductor Manufacturing International Corporation (SMIC), a state-backed Chinese foundry. This collaboration aimed to produce advanced chips using older equipment, a move fraught with uncertainty and high costs.\"}),/*#__PURE__*/e(\"p\",{children:\"By August 2023, despite the odds, Huawei launched the Mate 60 series phone, powered by the Kirin 9000S chip. This chip, although not as advanced as its contemporaries from Qualcomm, was well-received in China, marking Huawei's comeback in the chip market after years of sanctions. The US expressed surprise and concern over Huawei's ability to circumvent sanctions, sparking discussions about the implications of this technological breakthrough.\"}),/*#__PURE__*/e(\"p\",{children:\"The Kirin 9000S's development relied on less efficient deep ultraviolet (DUV) machines instead of the more advanced extreme ultraviolet (EUV) lithography. This choice was initially driven by budget constraints and later by necessity, as EUV equipment became inaccessible due to export controls. The use of DUV machines increased production steps, costs, and led to a lower yield rate (percentage of non-defective chips), but it was a critical factor in Huawei's ability to continue chip production.\"}),/*#__PURE__*/e(\"p\",{children:\"Significant state support and investment in Huawei and SMIC were evident. The Chinese government's financial backing played a crucial role in offsetting the high costs of chip production. This state support is part of China's broader strategy to achieve technological self-sufficiency and reduce reliance on foreign technology, especially given the ongoing geopolitical tensions and trade restrictions.\"}),/*#__PURE__*/e(\"p\",{children:\"Looking ahead, Huawei and SMIC are expanding their focus to include chips for artificial intelligence (AI) systems. This shift is in response to U.S. restrictions on high-performance chip sales to China. Huawei's Ascend series of AI chips have been identified as potential alternatives to Nvidia's products, despite lagging in overall performance. Huawei has already started trials of these chips with major Chinese internet companies including Tencent and Baidu.\"}),/*#__PURE__*/e(\"p\",{children:\"The success of the Kirin 9000S and the ongoing development of AI chips reflect China's determination to sustain and advance its semiconductor industry. However, challenges remain, particularly in the manufacturing process of data center chips, which are prone to a higher defect rate. Moreover, supply chain restrictions and the potential exhaustion of critical machine parts and materials pose additional hurdles for SMIC's production capabilities.\"}),/*#__PURE__*/e(\"p\",{children:\"The journey of Huawei and SMIC, supported by the Chinese state, underscores China's ambition to become a semiconductor superpower. It also highlights the complexities and interdependencies of the global tech industry, where geopolitical dynamics can significantly impact technological progress and market competition.\"})]}),[u]:\"\",[f]:!0,[w]:\"CTvKjSNWG\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"K39jrLMbz\",[c]:\"UAE's G42 Cuts Ties with Huawei\",[h]:\"uae-s-g42-cuts-ties-with-huawei\",[d]:\"2023-12-11T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/mM5Uknc9YvzPPHybJ8bmf4Bg.png\",srcSet:\"https://framerusercontent.com/images/mM5Uknc9YvzPPHybJ8bmf4Bg.png?scale-down-to=512 512w, https://framerusercontent.com/images/mM5Uknc9YvzPPHybJ8bmf4Bg.png 1024w\"},\"\"),[p]:\"UAE's G42 Cuts Ties with Huawei\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/i(\"p\",{children:[\"G42, the leading AI company out of the Middle East, that has been working with both Chinese and American tech giants, said it is cutting ties with Chinese tech giant Huawei. The move comes after \",/*#__PURE__*/e(\"em\",{children:\"The New York Times\"}),\" report on US government concerns about G42\u2019s relationships with the Chinese government and major tech companies.\"]}),/*#__PURE__*/i(\"p\",{children:[\"G42 CEO Peng Xiao told the \",/*#__PURE__*/e(r,{href:\"https://www.ft.com/content/6710c259-0746-4e09-804f-8a48ecf50ba3\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:/*#__PURE__*/e(\"em\",{children:\"Financial Times\"})})}),\" \u201CFor better or worse, as a commercial company, we are in a position where we have to make a choice. We cannot work with both sides. We can\u2019t.\u201D\"]}),/*#__PURE__*/e(\"p\",{children:\"G42 will be phasing out 5G infrastructure, servers, and data center tools from Huawei, which currently sit at the core of its infrastructure, in favor of US-made tech. Xiao also made it clear that G42 would work closely with the US to comply with export regulations on AI chips. Additionally, he emphasized that G42 never had \u201Cdeep AI research relationships\u201D with Chinese companies.\"}),/*#__PURE__*/e(\"p\",{children:\"However, industry leaders think that G42 completely severing ties with Chinese partners seems very unlikely. G42\u2019s technology developments and investments in AI are seen by many UAE officials as the country\u2019s future, potentially becoming one of its primary exports as they look to diversify from an oil-driven economy.\"}),/*#__PURE__*/e(\"p\",{children:\"G42, despite severing its connections with Huawei, continues to maintain partnerships with other significant Chinese entities. These include Sinopharm, the pharmaceutical company responsible for manufacturing and distributing the Covid vaccine in the UAE, and BGI, a genomics research organization that collaborated with G42 in establishing Covid detection centers. Subsidiaries of BGI were put on a Commerce Department blacklist by the Biden administration earlier this year for \u201Cposing a significant risk of contributing to monitoring and surveillance by the government of China.\u201D\"}),/*#__PURE__*/e(\"p\",{children:\"While G42 may be cozying up to the US now to get its hands on the best AI chips, the UAE doesn\u2019t want to put itself in a position where it is overly reliant on the US. It has been persistent in diversifying its technology partners, recently signing a deal with Swedish telecom giant Ericsson.\"}),/*#__PURE__*/e(\"p\",{children:\"This move reflects a broader strategy by the UAE to reduce reliance on any one country. Loyalty shouldn\u2019t be expected. China is the second biggest playing in the global AI arena. It would be difficult for any nation looking to become a leader in AI to do so without support from China\u2019s supply chain. While G42 has chosen a side today, expect alliances to shift over time as the UAE aims to bolster its position as a global power.\"})]}),[u]:\"\",[f]:!0,[w]:\"CTvKjSNWG\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"y37gaUsLw\",[c]:\"Abu Dhabi is Playing Both Sides with AI\",[h]:\"abu-dhabi-is-playing-both-sides-with-ai\",[d]:\"2023-12-04T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/Bd5XlmAgd80iuHKLgurhJfVTEvo.png\",srcSet:\"https://framerusercontent.com/images/Bd5XlmAgd80iuHKLgurhJfVTEvo.png?scale-down-to=512 512w, https://framerusercontent.com/images/Bd5XlmAgd80iuHKLgurhJfVTEvo.png 1024w\"},\"\"),[p]:\"Abu Dhabi is Playing Both Sides with AI\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/i(\"p\",{children:[\"Abu Dhabi-based G42 is the leading AI company out of the Middle East. The company is closely tied to the government with \",/*#__PURE__*/e(r,{href:\"https://en.wikipedia.org/wiki/Tahnoun_bin_Zayed_Al_Nahyan_(national_security_advisor)\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Tahnoun bin Zayed Al Nahyan\"})}),\"the, National Security Advisor of the UAE,\\xa0as the company\u2019s chairman and controlling shareholder.\"]}),/*#__PURE__*/i(\"p\",{children:[\"Recently, US officials have raised concerns that G42 is being used as a channel for China to get a hold of US technology and intelligence, and potentially even the genetic data of millions of Americans. \",/*#__PURE__*/e(r,{href:\"https://www.nytimes.com/2023/11/27/us/politics/ai-us-uae-china-security-g42.html\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/i(\"a\",{children:[/*#__PURE__*/e(\"em\",{children:\"The New York Times \"}),\"broke the bombshell story\"]})}),\" on White House concerns that the company may be feeding information to China and considers its ties to Chinese companies, most notably tech giant Huawei, which is at the core of G42 technology infrastructure, as a national security concern. Biden officials have put pressure on G42 leadership to cut ties with Chinese companies, which the firm has rejected. This summer the Biden administration held internal discussions about potentially sanctioning the firm.\"]}),/*#__PURE__*/e(\"p\",{children:\"The US has been investigating G42 CEO Peng Xiao, an American who renounced his citizenship to become a UAE citizen. The CIA issued a classified report on Xiao, which Xiao himself has said he\u2019s seen but \u201Chasn\u2019t finished reading\u201D. This isn\u2019t the first time Xiao has popped up on the radar of American intelligence agencies. In 2019, social media app ToTok, was identified as a spy tool for the Emirati government by US intelligence after it was found to be tracking the location and conversations of users. That data was being stored by UAE firm Pax AI, which Xiao controlled.\"}),/*#__PURE__*/i(\"p\",{children:[\"Following \",/*#__PURE__*/e(\"em\",{children:\"The Times \"}),\"report, which Xiao said he also hasn\u2019t read, he responded to data security concerns at \",/*#__PURE__*/e(r,{href:\"https://fortune.com/2023/11/29/g42-uae-abu-dhabi-ai-ceo-peng-xiao-new-york-times-report-china/\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Fortune Global Forum\"})}),\" in Abu Dhabi\",/*#__PURE__*/e(\"em\",{children:\" \"}),\"saying,\",/*#__PURE__*/e(\"em\",{children:\" \"}),\"\u201CWe make sure there is no leakage of sensitive information from our data center link in the U.S. If the U.S. counterparts ever share any data with us, we are accountable for ensuring that data is kept safe and sound here.\u201D\",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"But G42 has been playing both sides, working with both American and Chinese companies. In 2021, US private equity firm Silver Lake announced an $800m investment in G42. Since then, the company has inked deals across US industries - announcing strategic partnership with \",/*#__PURE__*/e(r,{href:\"https://www.g42.ai/resources/news/g42-healthcare-astrazeneca-enter-strategic-partnership\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"US pharmaceutical giant AstraZeneca\"})}),\", \",/*#__PURE__*/e(r,{href:\"https://www.arabianbusiness.com/industries/technology/abu-dhabis-g42-microsoft-to-offer-joint-sovereign-cloud-and-ai-capabilities\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Microsoft\"})}),\", \",/*#__PURE__*/e(r,{href:\"https://wam.ae/en/details/1395303113600\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Dell\"})}),\" and \",/*#__PURE__*/e(r,{href:\"https://www.prnewswire.com/news-releases/g42-and-openai-launch-partnership-to-deploy-advanced-ai-capabilities-optimized-for-the-uae-and-broader-region-301960623.html\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"OpenAI\"})}),\". It\u2019s also working with Silicon Valley chipmaker Cerebras, investing $100m to create the \u201Cworld\u2019s largest supercomputer\u201D.\"]}),/*#__PURE__*/e(\"p\",{children:\"In addition to its relationship with Huawei, G42 purchased $100m of shares in ByteDance, TikTok\u2019s parent company. Presight AI, a G42 subsidiary that sells surveillance technology to police forces worldwide, has come under the microscope of US officials who are concerned that the software looks nearly identical to surveillance tech that is popular with Chinese police. If Presight AI is selling white-labeled Chinese tech, it could give China unprecedented visibility to sensitive government data across the world.\"}),/*#__PURE__*/e(\"p\",{children:\"In light of the complex web of international concerns surrounding G42 and its CEO, it\u2019s evident that the company's actions and affiliations have triggered significant national security discussions. The ties between G42, Chinese entities, and the potential implications for US technology and intelligence have raised alarms. Despite the company's assurances of data security, questions linger about its dual engagements with American and Chinese firms. As this situation continues to unfold, the scrutiny on G42's activities underscores the importance of vigilance and transparency in the global technology landscape, as nations grapple with the challenges posed by the broader landscape of AI collaboration and data protection on the international stage.\"})]}),[u]:\"\",[f]:!0,[w]:\"ZUJZiBqb2\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"GBiuNyCqV\",[c]:\"Defense Using the Dark Arts: A Technical Intro to Adversarial ML\",[h]:\"defense-using-the-dark-arts-a-technical-intro-to-adversarial-ml\",[d]:\"2023-11-30T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/3McL8cKAcixvN6DeJ1DC9Qczc.png\",srcSet:\"https://framerusercontent.com/images/3McL8cKAcixvN6DeJ1DC9Qczc.png?scale-down-to=512 512w, https://framerusercontent.com/images/3McL8cKAcixvN6DeJ1DC9Qczc.png 1024w\"},\"\"),[p]:\"Defense Using the Dark Arts: A Technical Intro to Adversarial ML\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(\"em\",{children:\"This article was originally published in Black Box, the author's newsletter. Subscribe \"}),/*#__PURE__*/e(r,{href:\"https://jwang18.substack.com/\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:/*#__PURE__*/e(\"em\",{children:\"here\"})})}),/*#__PURE__*/e(\"em\",{children:\". \"})]}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"br\",{className:\"trailing-break\"})}),/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(\"em\",{children:\"This post assumes a conceptual understanding of neural network training, basic linear algebra, and familiarity with mathematical notation. For a refresher, see this \"}),/*#__PURE__*/e(r,{href:\"https://jwang18.notion.site/Gen-AI-Primer-4a014f7f6288429b9fa0f1a2a4fb7387?pvs=4\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:/*#__PURE__*/e(\"em\",{children:\"generative AI primer\"})})}),/*#__PURE__*/e(\"em\",{children:\" that I created.\"})]}),/*#__PURE__*/i(\"p\",{children:[\"I have been interested in protecting data and content from AI models since I proposed \",/*#__PURE__*/e(r,{href:\"https://jwang18.substack.com/p/hot-girl-singularity\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"accelerating model collapse\"})}),\" to preserve API access. I think that defensively using offensive techniques has a lot of potential, most of which is unrealized due to the nascency of this research. So when Nightshade, the \",/*#__PURE__*/e(r,{href:\"https://www.technologyreview.com/2023/10/23/1082189/data-poisoning-artists-fight-generative-ai/\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"data poisoning tool\"})}),\", was announced a few weeks ago, I was eager to dig into the paper.\\xa0\"]}),/*#__PURE__*/i(\"p\",{children:[\"What I quickly realized is in this branch of machine learning, there is a gap in knowledge between casual interest and understanding research. Luckily, I studied math alongside economics in college and had previously reviewed the \",/*#__PURE__*/e(r,{href:\"https://jwang18.notion.site/Gen-AI-Primer-4a014f7f6288429b9fa0f1a2a4fb7387?pvs=4\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"basics of generative AI\"})}),\". It was time for a deep dive.\\xa0\"]}),/*#__PURE__*/e(\"h3\",{children:\"Adversarial ML\"}),/*#__PURE__*/i(\"p\",{children:[\"Nightshade belongs to a field called \",/*#__PURE__*/e(r,{href:\"https://en.wikipedia.org/wiki/Adversarial_machine_learning\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"adversarial machine learning\"})}),\", which is the study of attacks on ML models and defenses against such attacks. To be specific, Nightshade is a data poisoning attack, which modifies the training data of a model so that it produces errors. This is related to another kind of attack called evasion, which modifies the inputs to make a model\u200A\u2014\u200Ausually a classifier\u200A\u2014\u200Asystematically produce wrong outputs, e.g., misclassify. Much adversarial ML research has been on evading image classifiers; Nightshade builds on this body of work.\"]}),/*#__PURE__*/e(\"p\",{children:\"Adversarial attacks can be further categorized as either\\xa0\"}),/*#__PURE__*/i(\"ul\",{style:{\"--framer-font-size\":\"19px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(64, 64, 64)\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"White box, in which the attacker has access to a model\u2019s parameters; or black box (more on that in a second).\\xa0\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Targeted, in which the attacker is aiming to produce errors of a specific class; or non-targeted. Most practical applications are targeted.\"})})]}),/*#__PURE__*/e(\"p\",{children:\"While the nature of adversarial spaces is unclear, the leading hypothesis is that they result from practical limitations. Since a model cannot be trained on the entire input space, the distributions of the training and test data will not match exactly. Adversarial space is the difference, and so inputs in that space are misclassified and become adversarial examples. \\xa0\"}),/*#__PURE__*/i(\"p\",{children:[\"Adversarial ML research has \u201Ctraditionally\u201D focused on white box attacks as they can transfer to black box models. This is possible since models trained on similar data are likely to have adversarial space that partially overlap. In fact, this probability increases with dimensionality due to a property called distance concentration\",/*#__PURE__*/e(r,{href:\"https://jwang18.substack.com/p/adversarial-ml#footnote-1-138896501\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"1\"})}),\". ML models are also inherently prone to behaving similarly since they are designed to generalize across training datasets and architecture. Similar classifiers should therefore partition their input space into classes with \",/*#__PURE__*/e(r,{href:\"https://en.wikipedia.org/wiki/Decision_boundary\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"decision boundaries\"})}),\" that are close to each other.\"]}),/*#__PURE__*/e(\"h3\",{children:\"Distance measures\"}),/*#__PURE__*/i(\"p\",{children:[\"What \u201Cclose\u201D means matters a lot in adversarial ML because attacks that are obviously manipulated will be caught. Instead of \",/*#__PURE__*/e(r,{href:\"https://en.wikipedia.org/wiki/Cosine_similarity\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"cosine similarity\"})}),\", the most common distance measures used in adversarial ML are \",/*#__PURE__*/e(r,{href:\"https://en.wikipedia.org/wiki/Lp_space#The_p-norm_in_finite_dimensions\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/i(\"a\",{children:[/*#__PURE__*/e(\"em\",{children:\"p\"}),\"-norms\"]})}),\", which are written as || \u2022 ||\",/*#__PURE__*/e(\"em\",{children:\"\u209A \"}),\"in literature. Given a pair of vectors, their\"]}),/*#__PURE__*/i(\"ul\",{style:{\"--framer-font-size\":\"19px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(64, 64, 64)\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(\"em\",{children:\"l\"}),\"\u2081\",/*#__PURE__*/e(\"em\",{children:\" \"}),\"norm or Manhattan distance is the sum of the absolute differences of their elements. It is often used for discrete data.\"]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(\"em\",{children:\"l\"}),\"\u2082 norm or Euclidean distance is given by the Pythagorean theorem. It is often used for continuous data.\\xa0\"]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(\"em\",{children:\"l\u209A\"}),\" norm or \",/*#__PURE__*/e(r,{href:\"https://en.wikipedia.org/wiki/Minkowski_distance\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Minkowski distance\"})}),\" generalizes this to vectors of any finite order \",/*#__PURE__*/e(\"em\",{children:\"p\"}),\".\\xa0\"]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(\"em\",{children:\"l\"}),\" \u035A\",/*#__PURE__*/e(\"em\",{children:\" \"}),\"norm or \",/*#__PURE__*/e(r,{href:\"https://en.wikipedia.org/wiki/Chebyshev_distance\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Chebyshev distance\"})}),\" is their largest difference among any of their dimensions.\"]})})]}),/*#__PURE__*/e(\"p\",{children:\"Speaking of literature, there are two foundational papers in adversarial ML that are worth understanding as background for Nightshade. I review these in turn.\\xa0\"}),/*#__PURE__*/e(\"h3\",{children:\"Fast Gradient Sign Method\\xa0(2015)\\xa0\"}),/*#__PURE__*/i(\"p\",{children:[\"An adversarial example takes advantage of the fact that a classifier \",/*#__PURE__*/e(\"em\",{children:\"C\"}),\" has a tolerance so that a perturbation \",/*#__PURE__*/e(\"em\",{children:\"\u03B7\"}),\" to an input \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" below a threshold \",/*#__PURE__*/e(\"em\",{children:\"\u03F5\"}),\" maps to the same class \",/*#__PURE__*/e(\"em\",{children:\"y\"}),\", i.e., \",/*#__PURE__*/e(\"em\",{children:\"C\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\") = \",/*#__PURE__*/e(\"em\",{children:\"C\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"x\u2019\"}),\") = \",/*#__PURE__*/e(\"em\",{children:\"y\"}),\" where \",/*#__PURE__*/e(\"em\",{children:\"x\u2019\"}),\" = \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" + \",/*#__PURE__*/e(\"em\",{children:\"\u03B7 \"}),\"and ||\",/*#__PURE__*/e(\"em\",{children:\"\u03B7\"}),\"|| \u035A < \",/*#__PURE__*/e(\"em\",{children:\"\u03F5\"}),\".\\xa0\"]}),/*#__PURE__*/i(\"p\",{children:[\"For intuition, consider the activation of a node in a neural network. Given a weight vector \",/*#__PURE__*/e(\"em\",{children:\"w\"}),\" and an adversarial example \",/*#__PURE__*/e(\"em\",{children:\"x\u2019\"}),\", the dot product is \",/*#__PURE__*/e(\"em\",{children:\"w\"}),\"\u1D40\",/*#__PURE__*/e(\"em\",{children:\"x\u2019\"}),\" = \",/*#__PURE__*/e(\"em\",{children:\"w\"}),\"\u1D40\",/*#__PURE__*/e(\"em\",{children:\"x + w\"}),\"\u1D40\",/*#__PURE__*/e(\"em\",{children:\"\u03B7\"}),\". The perturbation \",/*#__PURE__*/e(\"em\",{children:\"w\"}),\"\u1D40\",/*#__PURE__*/e(\"em\",{children:\"\u03B7\"}),\" can then be maximized\",/*#__PURE__*/e(r,{href:\"https://jwang18.substack.com/p/adversarial-ml#footnote-2-138896501\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"2\"})}),\" if \",/*#__PURE__*/e(\"em\",{children:\"\u03B7 \"}),\"= \",/*#__PURE__*/e(\"em\",{children:\"\u03F5sign\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"w\"}),\").\"]}),/*#__PURE__*/i(\"p\",{children:[\"This is a linear perturbation, but neural networks are generally nonlinear. However, \",/*#__PURE__*/e(r,{href:\"https://arxiv.org/pdf/1412.6572.pdf\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Goodfellow et al.\"})}),\" hypothesize that they behave linearly enough\u200A\u2014\u200Aat least locally\u200A\u2014\u200Athat they are susceptible to linear perturbations. (This can easily be shown, which they do in the rest of the paper.)\\xa0\"]}),/*#__PURE__*/i(\"p\",{children:[\"Their strategy is based on gradients. In training, backpropagation updates \",/*#__PURE__*/e(\"em\",{children:\"w\"}),\" to minimize the loss function \",/*#__PURE__*/e(\"em\",{children:\"J\"}),\" given an input \",/*#__PURE__*/e(\"em\",{children:\"x. \"}),\"A gradient-based attack is effectively the inverse as it holds \",/*#__PURE__*/e(\"em\",{children:\"w\"}),\" constant and perturbs \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" to maximize \",/*#__PURE__*/e(\"em\",{children:\"J\"}),\", subject to some \",/*#__PURE__*/e(\"em\",{children:\"\u03F5\"}),\".\\xa0\"]}),/*#__PURE__*/i(\"p\",{children:[\"For a model with parameters \",/*#__PURE__*/e(\"em\",{children:\"\u03B8\"}),\", the optimal perturbation is therefore\"]}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",\"data-framer-asset\":\"data:framer/asset-reference,OidRB9lujmYMCeoeSs1r6AuFM.jpg\",\"data-framer-height\":\"47\",\"data-framer-width\":\"356\",height:\"23\",src:\"https://framerusercontent.com/images/OidRB9lujmYMCeoeSs1r6AuFM.jpg\",style:{aspectRatio:\"356 / 47\"},width:\"178\"}),/*#__PURE__*/e(\"p\",{children:\"which Goodfellow et al. call the Fast Gradient Sign Method, a non-targeted white box attack. But what about targeted attacks?\"}),/*#__PURE__*/e(\"h3\",{children:\"Carlini and Wagner\\xa0(2017)\"}),/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(r,{href:\"https://arxiv.org/pdf/1608.04644.pdf\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Carlini and Wagner\"})}),\" propose the following to calculate a targeted white box attack on a classifier \",/*#__PURE__*/e(\"em\",{children:\"C\"}),\". Let \",/*#__PURE__*/e(\"em\",{children:\"C\"}),\"*(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\") be the correct class for a valid input \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" and \",/*#__PURE__*/e(\"em\",{children:\"t\"}),\" \u2260 \",/*#__PURE__*/e(\"em\",{children:\"C\"}),\"*(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\") be the target class. Finding an adversarial example \",/*#__PURE__*/e(\"em\",{children:\"x\u2019\"}),\" = \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" + \",/*#__PURE__*/e(\"em\",{children:\"\u03B4\"}),\" is then stated as \",/*#__PURE__*/e(\"em\",{children:\"min\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"D\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\", \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" + \",/*#__PURE__*/e(\"em\",{children:\"\u03B4\"}),\")) subject to \",/*#__PURE__*/e(\"em\",{children:\"C\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" + \",/*#__PURE__*/e(\"em\",{children:\"\u03B4\"}),\") = \",/*#__PURE__*/e(\"em\",{children:\"t\"}),\" and \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" + \",/*#__PURE__*/e(\"em\",{children:\"\u03B4\"}),\" \u2208 [0, 1]\",/*#__PURE__*/e(\"em\",{children:\"\u207F\"}),\", where \",/*#__PURE__*/e(\"em\",{children:\"D\"}),\" is a distance measure and \",/*#__PURE__*/e(\"em\",{children:\"n\"}),\" is the dimension of \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\".\"]}),/*#__PURE__*/i(\"p\",{children:[\"This is difficult to solve directly since \",/*#__PURE__*/e(\"em\",{children:\"C\"}),\" is nonlinear, so Carlini and Wagner define an objective function \",/*#__PURE__*/e(\"em\",{children:\"f\"}),\" such that \",/*#__PURE__*/e(\"em\",{children:\"C\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" + \",/*#__PURE__*/e(\"em\",{children:\"\u03B4\"}),\") = \",/*#__PURE__*/e(\"em\",{children:\"t \"}),\"if and only if\",/*#__PURE__*/e(\"em\",{children:\" f\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" + \",/*#__PURE__*/e(\"em\",{children:\"\u03B4\"}),\") \u2264 0. Conceptually, 1\u200A\u2014\u200A\",/*#__PURE__*/e(\"em\",{children:\"f\"}),\" is how close \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" + \",/*#__PURE__*/e(\"em\",{children:\"\u03B4 \"}),\"is to being classified as \",/*#__PURE__*/e(\"em\",{children:\"t, \"}),\"which makes \",/*#__PURE__*/e(\"em\",{children:\"f\"}),\" a loss function\",/*#__PURE__*/e(r,{href:\"https://jwang18.substack.com/p/adversarial-ml#footnote-3-138896501\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"3\"})}),\".\"]}),/*#__PURE__*/i(\"p\",{children:[\"If \",/*#__PURE__*/e(\"em\",{children:\"D\"}),\" is a \",/*#__PURE__*/e(\"em\",{children:\"p\"}),\"-norm, then the problem can be restated as \",/*#__PURE__*/e(\"em\",{children:\"min\"}),\"(||\",/*#__PURE__*/e(\"em\",{children:\"\u03B4\"}),\"||\",/*#__PURE__*/e(\"em\",{children:\"\u209A + cf\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" + \",/*#__PURE__*/e(\"em\",{children:\"\u03B4\"}),\")) subject to \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" + \",/*#__PURE__*/e(\"em\",{children:\"\u03B4\"}),\" \u2208 [0, 1]\",/*#__PURE__*/e(\"em\",{children:\"\u207F\"}),\", where \",/*#__PURE__*/e(\"em\",{children:\"c\"}),\" > 0 is a weighing factor for the loss term. Carlini and Wagner present several candidates for \",/*#__PURE__*/e(\"em\",{children:\"f\"}),\" and empirically find that the best is\\xa0\"]}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",\"data-framer-asset\":\"data:framer/asset-reference,WTI8JQWhyN9Cdba7PIyujz7yeY.jpg\",\"data-framer-height\":\"48\",\"data-framer-width\":\"524\",height:\"24\",src:\"https://framerusercontent.com/images/WTI8JQWhyN9Cdba7PIyujz7yeY.jpg\",style:{aspectRatio:\"524 / 48\"},width:\"262\"}),/*#__PURE__*/i(\"p\",{children:[\"where \",/*#__PURE__*/e(\"em\",{children:\"Z\"}),\"( \u2022 ) gives the output from all of a neural network\u2019s layers before the \",/*#__PURE__*/e(r,{href:\"https://en.wikipedia.org/wiki/Softmax_function\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"softmax function\"})}),\"\u200A\u2014\u200Aa generalization of the logistic function to \",/*#__PURE__*/e(\"em\",{children:\"n\"}),\" dimensions\u200A\u2014\u200Aand \",/*#__PURE__*/e(\"em\",{children:\"k\"}),\" is a confidence threshold (as a lower limit to the loss function). The \",/*#__PURE__*/e(\"em\",{children:\"Z\"}),\" expression is essentially the difference between what the classifier thinks \",/*#__PURE__*/e(\"em\",{children:\"x\u2019\"}),\" is and what the attacker wants the classifier to think it is\",/*#__PURE__*/e(r,{href:\"https://jwang18.substack.com/p/adversarial-ml#footnote-4-138896501\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"4\"})}),\".\"]}),/*#__PURE__*/i(\"p\",{children:[\"Carlini and Wagner make one more transformation to accommodate for the fact that many optimization algorithms are not bounded. They apply a change of variable\",/*#__PURE__*/e(r,{href:\"https://jwang18.substack.com/p/adversarial-ml#footnote-5-138896501\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"5\"})}),\" on \",/*#__PURE__*/e(\"em\",{children:\"\u03B4 \"}),\"and introduce \",/*#__PURE__*/e(\"em\",{children:\"w \"}),\"such that\",/*#__PURE__*/e(\"em\",{children:\" \u03B4 = \"}),\"\\xbd(\",/*#__PURE__*/e(\"em\",{children:\"tanh\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"w\"}),\") + 1)\u200A\u2014\u200A\",/*#__PURE__*/e(\"em\",{children:\"x.\"}),\" The final optimization problem is then\"]}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",\"data-framer-asset\":\"data:framer/asset-reference,5NRyRKWsLCgfaqLgWGTwYOk38.jpg\",\"data-framer-height\":\"86\",\"data-framer-width\":\"781\",height:\"43\",src:\"https://framerusercontent.com/images/5NRyRKWsLCgfaqLgWGTwYOk38.jpg\",style:{aspectRatio:\"781 / 86\"},width:\"390\"}),/*#__PURE__*/i(\"p\",{children:[\"which they solve using the \",/*#__PURE__*/e(r,{href:\"https://arxiv.org/pdf/1412.6980.pdf\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Adam optimizer\"})}),\". Carlini and Wagner\u2019s approach can quickly generate highly robust targeted white box attacks.\\xa0\"]}),/*#__PURE__*/e(\"h3\",{children:\"Glaze (2023)\"}),/*#__PURE__*/i(\"p\",{children:[\"Nightshade is an extension of prior work by Shan et al. that they call \",/*#__PURE__*/e(r,{href:\"https://arxiv.org/pdf/2302.04222.pdf\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Glaze\"})}),\", a style cloaking technique that applies Carlini and Wagner\u2019s optimization to text-to-image generators. These are harder to evade than image classifiers because they retain more of the features extracted from training images in order to generate original images\",/*#__PURE__*/e(r,{href:\"https://jwang18.substack.com/p/adversarial-ml#footnote-6-138896501\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"6\"})}),\":\"]}),/*#__PURE__*/i(\"ul\",{style:{\"--framer-font-size\":\"19px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(64, 64, 64)\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[\"During training, a text-to-image generator takes in an image \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" and uses a feature extractor \u03A6 to produce its feature vector \u03A6(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\").\\xa0\"]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[\"Simultaneously, an encoder \",/*#__PURE__*/e(\"em\",{children:\"E\"}),\" takes a corresponding text caption \",/*#__PURE__*/e(\"em\",{children:\"s\"}),\" and produces a predicted feature vector \",/*#__PURE__*/e(\"em\",{children:\"E\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"s\"}),\").\"]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[\"The parameters of \",/*#__PURE__*/e(\"em\",{children:\"E\"}),\" are optimized in training so that \",/*#__PURE__*/e(\"em\",{children:\"E\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"s\"}),\") = \u03A6(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\").\"]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[\"At generation time, a user passes a text prompt \",/*#__PURE__*/e(\"em\",{children:\"s\u1D62 \"}),\"into \",/*#__PURE__*/e(\"em\",{children:\"E\"}),\" and a decoder \",/*#__PURE__*/e(\"em\",{children:\"F\"}),\" decodes \",/*#__PURE__*/e(\"em\",{children:\"E\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"s\u1D62\"}),\") to produce the generated image \",/*#__PURE__*/e(\"em\",{children:\"x\u1D62\"}),\" = \",/*#__PURE__*/e(\"em\",{children:\"F\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"E\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"s\u1D62\"}),\")).\"]})})]}),/*#__PURE__*/e(\"p\",{children:\"Shan et al. focus on style mimicry attacks, where a text-to-image generator is used to create art in the style of a particular artist without their consent. Existing protective techniques rely on image cloaking, which are designed for classifiers; they are ineffective against style mimicry because they shift all features in an image instead of focusing on only features related to style.\"}),/*#__PURE__*/i(\"p\",{children:[\"Since it is difficult to explicitly identify and separate style features, Shan et al. do so implicitly by using another feature extractor \u03A9 to transfer artwork \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" into a target style \",/*#__PURE__*/e(\"em\",{children:\"T \"}),\"that is different from the artist\u2019s, \u03A9(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\", \",/*#__PURE__*/e(\"em\",{children:\"T\"}),\"). (\",/*#__PURE__*/e(\"em\",{children:\"T\"}),\" is selected based on the distances between the centroid of \u03A6(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\") and that of the feature spaces of candidate styles.) Then calculating the style cloak \",/*#__PURE__*/e(\"em\",{children:\"\u03B4\u2093 \"}),\"can be stated as the optimization \",/*#__PURE__*/e(\"em\",{children:\"min\"}),\"(\",/*#__PURE__*/e(\"em\",{children:\"D\"}),\"(\u03A6(\",/*#__PURE__*/e(\"em\",{children:\"x + \u03B4\u2093\"}),\")\u200A\u2014\u200A \u03A6(\u03A9(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\", \",/*#__PURE__*/e(\"em\",{children:\"T\"}),\")))) subject to |\",/*#__PURE__*/e(\"em\",{children:\"\u03B4\u2093| < p\"}),\", where \",/*#__PURE__*/e(\"em\",{children:\"D \"}),\"is a distance measure and \",/*#__PURE__*/e(\"em\",{children:\"p\"}),\" is the perturbation budget.\\xa0\"]}),/*#__PURE__*/e(\"p\",{children:\"Note that since Glaze is a black box evasion attack, the same model can act as both \u03A6 and \u03A9. In other words, Glaze can use DALL-E, Midjourney, Stable Diffusion, etc. against themselves! I found this to be particularly satisfying as it is the ultimate form of defensively using offensive techniques, at least from the perspective of the artists.\"}),/*#__PURE__*/e(\"p\",{children:\"Following Carlini and Wagner, Shan et al. then combines the restraint into the optimization problem\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",\"data-framer-asset\":\"data:framer/asset-reference,FcJrhajCDcVO3Jf9jsG6XajTRcM.jpg\",\"data-framer-height\":\"46\",\"data-framer-width\":\"915\",height:\"23\",src:\"https://framerusercontent.com/images/FcJrhajCDcVO3Jf9jsG6XajTRcM.jpg\",style:{aspectRatio:\"915 / 46\"},width:\"457\"}),/*#__PURE__*/i(\"p\",{children:[\"where \",/*#__PURE__*/e(\"em\",{children:\"\u03B1\"}),\" > 0 is a weighing factor for the loss term, \",/*#__PURE__*/e(r,{href:\"https://www.google.com/url?sa=i&url=https%3A%2F%2Farxiv.org%2Fpdf%2F1801.03924&psig=AOvVaw34OPj6jR-ZL9Lafu-Q6ojh&ust=1700854446119000&source=images&cd=vfe&ved=0CAgQrpoMahcKEwiItozj7tqCAxUAAAAAHQAAAAAQBA\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"LPIPS\"})}),\" is a measure of the perceived distortion, and \",/*#__PURE__*/e(\"em\",{children:\"D\"}),\" is instantiated as the \",/*#__PURE__*/e(\"em\",{children:\"l\"}),\"\u2082 norm.\"]}),/*#__PURE__*/i(\"p\",{children:[\"Glaze\u2019s style cloak was empirically successful in protecting art from being learned by Stable Diffusion and DALL-E, as judged by artists\",/*#__PURE__*/e(r,{href:\"https://jwang18.substack.com/p/adversarial-ml#footnote-7-138896501\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"7\"})}),\". It helps that artists are willing to accept fairly large \",/*#__PURE__*/e(\"em\",{children:\"p \"}),\"because their current methods of protecting are quite disruptive (e.g., watermarks). However, Glaze can only protect new art since most existing artwork is already part of training data.\"]}),/*#__PURE__*/e(\"h3\",{children:\"Nightshade (2023)\"}),/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(r,{href:\"https://arxiv.org/pdf/2310.13828.pdf\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Nightshade\"})}),\" goes beyond cloaking and damages the offending model itself. It takes advantage of the fact that text-to-image generators exhibit concept sparsity despite having large training datasets. That is, a very small portion of the training data contains a given term or its semantically related terms. As a result, Shan et al. hypothesize that text-to-image generators are much more vulnerable to data poisoning (at the concept level) than is commonly believed.\"]}),/*#__PURE__*/e(\"p\",{children:\"They prove this by proposing Nightshade, a prompt-specific data poisoning attack based on mismatched text/image pairs. To minimize the number of poison samples for ease of implementation, Nightshade follows two design principles:\\xa0\"}),/*#__PURE__*/i(\"ul\",{style:{\"--framer-font-size\":\"19px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(64, 64, 64)\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[\"Maximize the effect of each sample by including the keyword \",/*#__PURE__*/e(\"em\",{children:\"C\"}),\" in each poison prompt so that it targets only the parameters associated with \",/*#__PURE__*/e(\"em\",{children:\"C\"}),\".\\xa0\"]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[\"Minimize conflicts among different pairs, and thus the overlap of their contributions to the perturbation norm, by creating original images of the target concept \",/*#__PURE__*/e(\"em\",{children:\"T \"}),\"using another\",/*#__PURE__*/e(\"em\",{children:\" \"}),\"text-to-image generator.\"]})})]}),/*#__PURE__*/i(\"p\",{children:[\"If \u03A6 is the feature extractor of the victim model and \u03A9 is that of the poison image model, than a valid image \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" corresponding to the poison prompt can be perturbed by \",/*#__PURE__*/e(\"em\",{children:\"\u03B4\u2093\"}),\" so that it is poisoned into \u03A9(\",/*#__PURE__*/e(\"em\",{children:\"x\"}),\", \",/*#__PURE__*/e(\"em\",{children:\"T\"}),\"). This can be calculated using Glaze\",/*#__PURE__*/e(r,{href:\"https://jwang18.substack.com/p/adversarial-ml#footnote-8-138896501\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"8\"})}),\".\\xa0\"]}),/*#__PURE__*/e(\"p\",{children:\"Nightshade produces stronger poisoning effects than previous techniques and they bleed through to related concepts. They also stack if concepts are combined into a single prompt. Furthermore, if many Nightshade attacks target different prompts on the same model, general features will become corrupted and its image generation function eventually collapses!\"}),/*#__PURE__*/e(\"p\",{children:\"It is not difficult to imagine that all platforms that host media will protect their content with adversarial techniques like these in the future. I would love to learn more if you\u2019re building or researching in this space\u2014reach out at wngjj[dot]61[at]gmail.com. Until then, I will be thinking about how every new offensive threat could be a new defensive opportunity. \u220E\"}),/*#__PURE__*/i(\"ol\",{children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Volume generalizes as distance raised to the power of the dimension, so increasingly more volume is on the edges at high dimensions. Here, the edges are decision boundaries.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[\"This can be arbitrarily large for sufficiently high \",/*#__PURE__*/e(\"em\",{children:\"n\"}),\" since \",/*#__PURE__*/e(\"em\",{children:\"l\"}),\" \u035A does not grow with dimensionality.\"]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[\"Defining \",/*#__PURE__*/e(\"em\",{children:\"f\"}),\" as the complement probability also enables it to be combined with \",/*#__PURE__*/e(\"em\",{children:\"D\"}),\" into a single minimization in the next step.\"]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[\"The first term represents the probability of the class of which \",/*#__PURE__*/e(\"em\",{children:\"C\"}),\" predicts \",/*#__PURE__*/e(\"em\",{children:\"x'\"}),\" is part and the second term represents the probability that \",/*#__PURE__*/e(\"em\",{children:\"C\"}),\" classifies \",/*#__PURE__*/e(\"em\",{children:\"x'\"}),\" as \",/*#__PURE__*/e(\"em\",{children:\"t\"}),\" instead. Note that \",/*#__PURE__*/e(\"em\",{children:\"Z\"}),\" is not a probability itself but a softmax input.\"]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[\"Since \",/*#__PURE__*/e(\"em\",{children:\"tanh\"}),\"( \u2022 ) \u2208 [-1, 1], this is equivalent to \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\" + \",/*#__PURE__*/e(\"em\",{children:\"\u03B4\"}),\" \u2208 [0, 1]\",/*#__PURE__*/e(\"em\",{children:\"\u207F\"}),\".\"]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"This is a very large output space. Classifiers have limited output classes and therefore do not have to retain as many features.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Shan et al. worked closely with professional artists and the Glaze paper includes their perspectives in \\xa73.1, which I recommend reading.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[\"In practice, Nightshade uses prompts from a valid dataset of text/image pairs to easily find \",/*#__PURE__*/e(\"em\",{children:\"x\"}),\".\"]})})]})]}),[u]:\"Jiarui Wang\",[f]:!0,[w]:\"CTvKjSNWG\",[y]:!0,[b]:\"sDUwRkrv2\",[v]:\"X8bGnpFRB\"},{id:\"e5UiXay76\",[c]:\"China's AI Smart Cities\",[h]:\"china-s-ai-smart-cities\",[d]:\"2023-11-27T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/zswatuu1AjzVQ33SlrwGZfeCV2M.png\",srcSet:\"https://framerusercontent.com/images/zswatuu1AjzVQ33SlrwGZfeCV2M.png?scale-down-to=512 512w, https://framerusercontent.com/images/zswatuu1AjzVQ33SlrwGZfeCV2M.png 1024w\"},\"\"),[p]:\"China's AI Smart Cities\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Beijing, Shanghai, Guangzhou, Shenzhen, and Hangzhou - China\u2019s smart cities that are rapidly integrating AI and IoT (Internet of Things) into public services and everyday life are setting the standard for the world in what the cities of the future will look like.\"}),/*#__PURE__*/e(\"p\",{children:\"In what sounds straight out of a science fiction movie, at new subway stations in Beijing passengers no longer need to swipe a card or tap their phones, instead facial recognition software scans their face to verify their identity and process payment. Large screens in stations show real-time capacity in each carriage to let passengers know which to board.\"}),/*#__PURE__*/e(\"p\",{children:\"This is a part of the Beijing Citizen Social Service Card which has turned the city effectively cashless, where citizens can automatically \u201Cpay with their faces\u201D at supermarkets and restaurants, for parking spots, and even hospital visits digitally after they have received their goods or services.\"}),/*#__PURE__*/e(\"p\",{children:\"Shanghai\u2019s Citizen Cloud offers over 1,000 different public services for residents such as storage of important personal documents, transportation, and leisure activities. Some techno-optimists may consider this a process made more efficient with technology, while alarmists would find this concerning, just a few steps away from a society without privacy.\"}),/*#__PURE__*/i(\"p\",{children:[\"No matter where you stand, this present day scenario in China is likely to come to the US as technology advances and AI becomes more integrated into public infrastructure. \",/*#__PURE__*/e(r,{href:\"https://fedscoop.com/us-spending-on-ai-hit-3-3b-in-fiscal-2022/\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"The US government spent $3.3B on AI last year\"})}),\", up 2.5x in the last 5 years, a trend expected to continue to grow. It\u2019s worth looking at what China is doing with AI to create smart cities to understand what to expect in the future for US cities and how we could learn how to replicate successes and avoid potential issues that could jeopardize American freedoms.\"]}),/*#__PURE__*/e(\"p\",{children:\"Americans have been introduced to these types of technologies in the home - with security systems and home appliances. In China these new technologies are rapidly being deployed into major cities. The city of Guangzhou for example, uses sensor technology to monitor the moisture level in the soil and get a feed of data on the health of the trees and heavy foliage in the city. This allows for better care, practically eliminating the eyesore of dead plants that plague many cities - creating a future of more beautiful cityscapes. Similar technology is implemented to monitor air quality and pollutants in Guangzhou and other major cities.\"}),/*#__PURE__*/e(\"p\",{children:\"The city of Hangzhou, the home of tech giant Alibaba, has implemented Alibaba\u2019s ET City Brain. The technology monitors traffic conditions in real-time to adjust traffic lights. Since 2016 the city has seen an overall reduction of travel time by three minutes and a 50% faster response time for emergency vehicles. The City Brain system has been implemented in 23 cities across Asia, while successful in reducing traffic congestion, it isn\u2019t without criticism. The technology was widely adopted across China during the pandemic for surveillance, to monitor citizens and alert authorities of large gatherings, which many have called a government overreach and invasion of privacy.\"}),/*#__PURE__*/e(\"p\",{children:\"As we observe the evolution of China's smart cities, where AI and IoT are seamlessly woven into the urban fabric, we're witnessing the blueprint of future metropolises. As the US ramps up its investment in AI, it must find a balance in leveraging technology for public good and safeguarding individual freedoms. This will be a defining challenge as American cities inch closer to their own smart futures.\"})]}),[u]:\"\",[f]:!0,[w]:\"I8_efFBQ0\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"zCWHuUYoc\",[c]:\"Biden & Xi Jinping APEC Summit Meeting Recap\",[h]:\"biden-xi-jinping-apec-meeting-recap\",[d]:\"2023-11-20T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/oWgBW0yzOTrGkenYLnKHqC4XU4.png\",srcSet:\"https://framerusercontent.com/images/oWgBW0yzOTrGkenYLnKHqC4XU4.png?scale-down-to=512 512w, https://framerusercontent.com/images/oWgBW0yzOTrGkenYLnKHqC4XU4.png 1024w\"},\"\"),[p]:\"Biden & Xi Jinping APEC Summit Meeting Recap\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"President Biden and Xi Jinping had their first face-to-face meeting in over a year last Wednesday. It was a productive meeting that yielded agreements to resume military communications, which China cut off last August after Nancy Pelosi\u2019s visit to Taiwan, and open up a presidential hotline between the two leaders for immediate direct communication at the highest level.\"}),/*#__PURE__*/e(\"p\",{children:\"Despite reports from both US and Chinese news outlets last week on a potential agreement to limit the use of AI in nuclear weapons, there was no formal agreement made. The conversation on AI was underwhelming. The leaders did discuss AI, releasing a statement after the meeting that they agreed to \u201Caddress the risks of advanced AI systems and improve AI safety through U.S.-China government talks.\u201D No additional details on the AI discussions have been released.\"}),/*#__PURE__*/i(\"p\",{children:[\"Biden didn\u2019t back down on US sanctions blocking the export of powerful semiconductor chips used for AI development, saying that they are not likely to change any time soon. \",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"Ian Bremmer, head of global politics research firm Eurasia Group, spoke at the summit warning political leaders in the audience: \u201CThe United States and China for now appear to be heading toward a technology cold war. That has not been fixed, that has not been addressed.\u201D\",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"Still, shares of major US tech companies like Nvidia and Google rose late last week signaling optimism from the tech community as a result of the meeting.\"]}),/*#__PURE__*/e(\"p\",{children:\"There were positive outcomes in non-tech diplomacy. Xi agreed to go after Chinese chemical companies that are exporting fentanyl-related products to the US & Mexico, one of Biden\u2019s top priorities going into the meeting to curb the opioid crisis in America.\"}),/*#__PURE__*/e(\"p\",{children:\"Biden pushed for \u201Cpeace and stability\u201D in Taiwan while Xi agreed that this was important for the short-term, there needed to be a \u201Cpeaceful resolution\u201D for the reunification of Taiwan as a part of China.\"}),/*#__PURE__*/e(\"p\",{children:\"When asked if he still viewed Xi Jinping as a dictator, Biden responded:\"}),/*#__PURE__*/e(\"p\",{children:'\"Well, look, he is. I mean, he\u2019s a dictator in the sense that he is a guy who runs a country that is a communist country,\" Biden said.'}),/*#__PURE__*/e(\"p\",{children:\"Overall, the meeting was fruitful for US-China relations, and a relief for both countries as tensions have been rising over the last year. Developing the best tech will continue to be a top priority for both countries as we head into an AI cold war, but for now, the communication lines are open and tensions have slightly diffused.\"})]}),[u]:\"AMID AI\",[f]:!0,[w]:\"I8_efFBQ0\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"ziPREn_5N\",[c]:\"Biden & Xi Jinping Set to Meet\",[h]:\"biden-xi-jinping-set-to-meet\",[d]:\"2023-11-13T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/H0ZxxbzP4EzzCRtVDxsfK28bZY4.png\",srcSet:\"https://framerusercontent.com/images/H0ZxxbzP4EzzCRtVDxsfK28bZY4.png?scale-down-to=512 512w, https://framerusercontent.com/images/H0ZxxbzP4EzzCRtVDxsfK28bZY4.png 1024w\"},\"\"),[p]:\"Biden & Xi Jinping Set to Meet\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/i(\"p\",{children:[\"President Biden and Chinese leader Xi Jinping are expected to have their first face-to-face meeting in over a year this Wednesday after the APEC summit in San Francisco. It will be a closed door meeting, the time and location have not been disclosed for security reasons - Chinese officials have also urged the US to ensure that there are no protestors anywhere near Xi Jinping\u2019s motorcade or hotel. \",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"Tensions between the two superpowers have escalated over the last year, specifically around technology, US support of Taiwan, and China\u2019s support for Russia in the Ukraine war. Tensions are high on a personal level as well, after Biden publicly called Xi Jinping \u201Ca dictator\u201D in June.\"]}),/*#__PURE__*/e(\"p\",{children:\"One of the primary goals of the meeting will be to reestablish communication channels between militaries, which China cut off in August of last year in response to Nancy Pelosi\u2019s visit to Taiwan. Military tensions have escalated in recent months as China has been moving warships and air crafts around the South China Sea, in what the US military is calling a show of force. Just a few weeks ago, a Chinese fighter jet risked collision as it flew within 10 feet of a US bomber at high speed.\"}),/*#__PURE__*/i(\"p\",{children:[\"Both Chinese and US officials don\u2019t expect any breakthrough or lasting agreements to come from the meeting. It\u2019s expected to be a showcase of cooperation with asks from both sides.\",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"Along with reestablishing military communication and deescalation in the South China Sea, Biden will be looking for assurances from China in cutting off it\u2019s chemical exports to Mexico, which have been fueling the fentanyl crisis in the US.\"]}),/*#__PURE__*/i(\"p\",{children:[\"China will be looking for Biden to commit that the US will not support Taiwan\u2019s formal independence, and more accommodations to recognize and help China grow as a rising global power (by reducing technology sanctions), which isn\u2019t likely. \",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"However, coming out of Chinese media, \",/*#__PURE__*/e(\"em\",{children:\"South China Morning Post\"}),\" \",/*#__PURE__*/e(r,{href:\"https://www.scmp.com/news/china/military/article/3241177/biden-xi-set-pledge-ban-ai-autonomous-weapons-drones-nuclear-warhead-control-sources\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"reported that\"})}),\" Biden and Xi Jinping are expected to set a landmark agreement that pledges \u201Ca ban on the use of\\xa0artificial intelligence\\xa0in autonomous weaponry, such as drones, and in the control and deployment of nuclear warheads,\u201D. This information has yet to be confirmed by any US-based news outlets.\"]})]}),[u]:\"\",[f]:!0,[w]:\"ZUJZiBqb2\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"xizCynhbK\",[c]:\"Biden's AI Executive Order\",[h]:\"bidens-ai-executive-order\",[d]:\"2023-11-06T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/5DwVB1vPhe7s4Xdv0OrFIjqX1k.png\",srcSet:\"https://framerusercontent.com/images/5DwVB1vPhe7s4Xdv0OrFIjqX1k.png?scale-down-to=512 512w, https://framerusercontent.com/images/5DwVB1vPhe7s4Xdv0OrFIjqX1k.png 1024w\"},\"\"),[p]:\"Biden's AI Executive Order\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/i(\"p\",{children:[\"Last Monday Biden invoked the Korean War-era Defense Production Act to push through the \",/*#__PURE__*/e(\"em\",{children:\"Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence\"}),\", which requires AI companies to notify the government when developing technology or systems that pose \u201Cserious risk to national security, national economic security or national public health and safety,\u201D\"]}),/*#__PURE__*/i(\"p\",{children:[\"The \",/*#__PURE__*/e(r,{href:\"https://www.whitehouse.gov/briefing-room/presidential-actions/2023/10/30/executive-order-on-the-safe-secure-and-trustworthy-development-and-use-of-artificial-intelligence/\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"order\"})}),\" announced eight guiding principles and priorities for AI:\"]}),/*#__PURE__*/i(\"ol\",{children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Artificial Intelligence must be safe and secure.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Responsible innovation, competition, and collaboration will allow the United States to lead in AI and unlock the technology\u2019s potential to solve some of society\u2019s most difficult challenges.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"The responsible development and use of AI require a commitment to supporting American workers.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Artificial Intelligence policies must be consistent with the Biden Administration\u2019s dedication to advancing equity and civil rights.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"The interests of Americans who increasingly use, interact with, or purchase AI and AI-enabled products in their daily lives must be protected.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Americans\u2019 privacy and civil liberties must be protected as AI continues advancing.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"It is important to manage the risks from the Federal Government\u2019s own use of AI and increase its internal capacity to regulate, govern, and support responsible use of AI to deliver better results for Americans.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"The Federal Government should lead the way to global societal, economic, and technological progress, as the United States has in previous eras of disruptive innovation and change.\"})})]}),/*#__PURE__*/e(\"p\",{children:\"The order also requires US AI firms to notify the government within 90 days if a foreign company uses their platforms to train extensive AI systems - which is mainly aimed at foreign adversaries like China and Russia who may use US tech to create disinformation campaigns with deepfakes and develop military technologies.\"}),/*#__PURE__*/i(\"p\",{children:[\"Straight out of the movie \",/*#__PURE__*/e(\"em\",{children:\"Minority Report\"}),\", the order included the following language:\"]}),/*#__PURE__*/e(\"p\",{children:\"\u201Cwithin 365 days of the date of this order, submit to the President a report that addresses the use of AI in the criminal justice system, including any use in:\\xa0crime forecasting and predictive policing, including the ingestion of historical crime data into AI systems to predict high-density \u2018hot spots\u2019.\u201D\"}),/*#__PURE__*/e(\"p\",{children:\"The White House called the order \u201Cthe most significant actions ever taken by any government to advance the field of AI safety.\u201D The order was released just before the UK AI Safety Summit, which was attended by the US and China - likely to make a statement that the US is leading the way in AI regulation and set a standard for other countries to follow.\"}),/*#__PURE__*/i(\"p\",{children:[\"There has been some push back to the order. Prominent ventures capitalists, including Marc Andreessen, \",/*#__PURE__*/e(r,{href:\"https://x.com/martin_casado/status/1720517026538778657?s=20\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"submitted a letter\"})}),\" to President Biden calling the new guidelines too restricting on open-source AI software - which they see as pivotal to a free and safe world in the age of AI.\"]}),/*#__PURE__*/e(\"p\",{children:\"MIT Professor Max Tegmark addressed Congress last week about the existential threats posed by superintelligent AI, emphasizing the need for swift regulatory action. While he applauded the order, he said it wasn\u2019t enough, emphasizing the need for Congress to pass actual laws, while highlighting concerns about both China\u2019s AI ambitions and the threat of creating \u2018superintelligence\u2019 in the US that would \u2018make humans completely obsolete\u2019.\"}),/*#__PURE__*/e(\"p\",{children:\"But AI companies appear to be falling in line. To close out the UK\u2019s AI Safety Summit last Thursday, companies including OpenAI, Google, Anthropic, Amazon, Microsoft, and Meta signed a (non-legally binding) letter agreeing to allow governments including the US, UK, and Singapore to test their models for national security and safety risks. China and Chinese tech companies notably did not sign the letter.\"})]}),[u]:\"\",[f]:!0,[w]:\"I8_efFBQ0\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"cmjmvpigU\",[c]:\"The 20 Best Artificial Intelligence Books to Read Now\",[h]:\"the-20-best-artificial-intelligence-books-to-read-now\",[d]:\"2023-11-08T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/yeU20bwnJvx1TyD8C0DB3rcgYMc.png\",srcSet:\"https://framerusercontent.com/images/yeU20bwnJvx1TyD8C0DB3rcgYMc.png?scale-down-to=512 512w, https://framerusercontent.com/images/yeU20bwnJvx1TyD8C0DB3rcgYMc.png 834w\"},\"\"),[p]:\"The 20 Best Artificial Intelligence Books to Read Now\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"ol\",{children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Superintelligence-Dangers-Strategies-Nick-Bostrom/dp/0198739834/ref=pd_lpo_sccl_2/144-0250447-2632956?pd_rd_w=jNcxV&content-id=amzn1.sym.116f529c-aa4d-4763-b2b6-4d614ec7dc00&pf_rd_p=116f529c-aa4d-4763-b2b6-4d614ec7dc00&pf_rd_r=ZN37ECZFMESP52ARXZ9F&pd_rd_wg=lTBkR&pd_rd_r=dc45a88b-0a58-4ea8-9d3b-7498bd11616f&pd_rd_i=0198739834&psc=1\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Superintelligence: Paths, Dangers, Strategies\\xa0\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Nick Bostrom\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/9c19b950-7e51-4e94-8720-e418e00b30e6\"}),/*#__PURE__*/e(\"p\",{children:\"A philosophical exploration of the potential rise of superintelligence, discussing the paths to its creation, the inherent dangers, and possible strategies for managing its development. Bostrom contemplates the control problem of ensuring a superintelligence's goals are aligned with human well-being, highlighting the difficulty of this task due to the unpredictable nature of such an advanced entity. The book serves as a cautionary tale, stressing the urgency of addressing the existential risks of AI before they become unmanageable\"}),/*#__PURE__*/e(\"ol\",{start:\"2\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Hundred-Page-Machine-Learning-Book/dp/199957950X\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"The Hundred-Page Machine Learning Book\\xa0\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Andriy Burkov\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/270b0b3a-13d7-4b8a-9bb1-e0251fdf76cc\"}),/*#__PURE__*/e(\"p\",{children:\"A concise and efficient overview of machine learning (ML) concepts, aimed at a broad audience of data science enthusiasts with varying levels of expertise. It's a dense read that efficiently covers complex topics, extending beyond the basics to advanced ML models and new research areas like zero-shot learning. While accessible, the book challenges readers to deepen their understanding of ML, regardless of whether they're undergraduates, data professionals, or seasoned data scientists\u200B.\"}),/*#__PURE__*/e(\"ol\",{start:\"3\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/AI-2041-Ten-Visions-Future/dp/059323829X\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"AI 2041: Ten Visions for Our Future B\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Kai-Fu Lee & Chen Qiufan\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/e8333f1c-7322-4730-9d98-3eaa580272cb\"}),/*#__PURE__*/e(\"p\",{children:\"This book merges storytelling and scientific prediction to explore how AI might transform our world in the next two decades. The authors, an AI expert and a speculative fiction writer, use short stories to illustrate possible futures shaped by AI advancements. They predict significant changes in work, communication, and social norms, while also acknowledging potential risks like autonomous weapons and biases in smart technologies. The book emphasizes that despite AI's transformative impact, humanity will ultimately control its own destiny.\"}),/*#__PURE__*/e(\"ol\",{start:\"4\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/AI-Superpowers-China-Silicon-Valley/dp/132854639X\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"AI Superpowers: China, Silicon Valley, and the New World Order\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Kai-fu Lee\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/645bfd64-005a-4d47-9d52-35ee81c517ba\"}),/*#__PURE__*/e(\"p\",{children:\"Another gem by Kai-Fu Lee. AI Superpowers examines the AI industry's growth, particularly in China, and its implications on global power dynamics and employment. Lee explores how China's unique approach to AI development, with fewer data protections and an aggressive tech culture, positions it as a leader in the global AI race\"}),/*#__PURE__*/e(\"ol\",{start:\"5\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Our-Final-Invention-Artificial-Intelligence/dp/0312622376\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Our Final Invention: Artificial Intelligence and the End of the Human Era\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By James Barrat\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/0af34a9c-bb87-44f4-a611-12a9a699ac65\"}),/*#__PURE__*/e(\"p\",{children:\"A critical look at the future of artificial intelligence and its potential to surpass human intelligence. The book explores AI concepts and history, and features insights from notable AI researchers. Barrat warns of the existential risks posed by super-intelligent AI, including the potential for human extinction. He emphasizes the challenge in predicting and controlling an entity that might far exceed human intelligence, urging caution as we advance in AI development\u200B.\"}),/*#__PURE__*/e(\"ol\",{start:\"6\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/If-Then-Company-Invented-Future/dp/1529386152\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"If Then: How One Data Company Invented the Future\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Jill Lepore\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/bba9d3b6-c467-45ae-9972-00ff2604cec8\"}),/*#__PURE__*/e(\"p\",{children:\"A historical examination by Jill Lepore of the Simulmatics Corporation. Founded in 1959, this company was a precursor to today's tech giants, pioneering the use of computer simulations to predict and influence human behavior for purposes ranging from marketing to political campaigns. Despite its ambitious start and involvement in significant events like the JFK campaign and the Vietnam War, Simulmatics collapsed amidst controversy. Lepore's narrative explores the early ambitions and ultimate failures of the company, reflecting on the power and danger of data and predictive modeling long before modern data giants emerged.\"}),/*#__PURE__*/e(\"ol\",{start:\"7\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Algorithmic-Leader-Smart-Machines-Smarter/dp/1989025331\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"The Algorithmic Leader\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Mike Walsh\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/79126cde-d94b-415c-894d-815e61747940\"}),/*#__PURE__*/e(\"p\",{children:\"A visionary book outlining how to excel in an era where AI and automation are pervasive. Walsh, a futurist, argues that the key challenge is not the threat of robots taking over jobs, but rather our capacity to reinvent ourselves in the face of these technological advances. Through interviews with business leaders and AI experts, Walsh has distilled 10 core principles for success in the algorithmic age, focusing on embracing change and leveraging the power of algorithms to enhance human decision-making and creativity.\\xa0\"}),/*#__PURE__*/e(\"ol\",{start:\"8\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/You-Look-Like-Thing-Love/dp/0316525227\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"You Look Like a Thing and I Love You: How Artificial Intelligence Works and Why It's Making the World a Weirder Place\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Janelle Shane\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/cccaa34c-5f2a-4f11-ad3e-718b352f3464\"}),/*#__PURE__*/e(\"p\",{children:\"Shane, an optics research scientist, uses humor and accessible language to explain how AI works and its impact on society. The title itself is a humorous nod to the unexpected outputs AI can generate. Throughout the book, Shane addresses the concept of \\\"artificial stupidity,\\\" highlighting the limitations and oddities of AI through various anecdotes and principles, such as AI's tendency to take shortcuts and its lack of general intelligence. Critics have praised the book as eye-opening, fun, and a great guide for understanding AI's oddities without the jargon.\"}),/*#__PURE__*/e(\"ol\",{start:\"9\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Reality-Game-Technology-Break-Truth/dp/1541768256\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"The Reality Game: How the Next Wave of Technology Will Break the Truth\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Samuel Woolley\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/52d0f4dd-abf2-4fa8-b802-f4ba428ee8ec\"}),/*#__PURE__*/e(\"p\",{children:\"A forward-looking examination of how emerging technologies, especially in social media, could undermine the truth and influence our perceptions and beliefs. Woolley, focusing on the social, economic, and political factors driving the manipulative uses of technology, emphasizes the need to address these underlying issues through a lens of diplomacy and human rights, not just as warfare. The book stresses the role of humans behind the technology and urges action to prevent the misuse of future technologies, aiming to support democracy rather than diminish it. Woolley adopts a cautiously optimistic approach, suggesting that while the technology itself is not malevolent, its potential misuse necessitates proactive solutions.\"}),/*#__PURE__*/e(\"ol\",{start:\"10\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Big-Nine-Thinking-Machines-Humanity/dp/1541773756\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"The Big Nine: How the Tech Titans and Their Thinking Machines Could Warp Humanity\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Amy Webb\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/a608f17c-b9f6-4e0c-8372-9d82a001b714\"}),/*#__PURE__*/e(\"p\",{children:\"A critical examination of AI's potential trajectory, shaped by the major tech corporations\u2014Amazon, Google, Facebook, Tencent, Baidu, Alibaba, Microsoft, IBM, and Apple. Webb argues that these entities, driven by profit rather than communal well-being, may inadvertently create AI systems that act unpredictably and contrary to human values. The book is a call to redirect the course of AI development towards a future aligned with human interests and liberate society from the unintended consequences of algorithmic decision-making.\"}),/*#__PURE__*/e(\"ol\",{start:\"11\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Futureproof-Rules-Humans-Age-Automation/dp/059313334X\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Futureproof: 9 Rules for Humans in the Age of Automation\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Kevin Roose\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/60d44d51-4689-481e-be6d-2413dc1126d8\"}),/*#__PURE__*/e(\"p\",{children:\"Roose, a New York Times technology columnist, provides a pragmatic guide for ensuring that humans remain relevant by enhancing qualities that machines cannot replicate. His approach emphasizes the importance of being surprising, social, and scarce; resisting the passive adoption of technology ('machine drift'); and using AI as a tool without becoming machine-like ourselves\"}),/*#__PURE__*/e(\"ol\",{start:\"12\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Human-Algorithm-Artificial-Intelligence-Redefining/dp/1640092366\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"A Human Algorithm: How Artificial Intelligence Is Redefining Who We Are\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Flynn Coleman\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/88515810-adf0-466b-9b46-556605016ef7\"}),/*#__PURE__*/e(\"p\",{children:\"This book discusses the imperative of embedding human values into AI to manage the power shift from humans to machines. Coleman posits that AI can enhance the human condition if it's developed ethically. The book highlights the need for oversight mechanisms and readiness for more advanced AI forms. The book has been described as substantial and energetic, though its optimism was met with some skepticism from prominent AI researchers.\\xa0\"}),/*#__PURE__*/e(\"ol\",{start:\"13\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Deep-Thinking-Machine-Intelligence-Creativity/dp/161039786X\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Deep Thinking: Where Machine Intelligence Ends and Human Creativity Begins\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Garry Kasparov, Mig Greengard\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/c22d9e0c-1483-4746-886c-95023fd6e960\"}),/*#__PURE__*/e(\"p\",{children:\"Co-authored by Garry Kasparov, one of the preeminent chess grandmasters of all time, the book provides an in-depth look into the evolution of machine intelligence in the realm of chess, from Kasparov's historic 1997 match against IBM's Deep Blue to the broader implications for human creativity and future technology. Kasparov shares his unique perspective on this pivotal moment in AI development and reflects on the interplay between human and machine intelligence, offering insights into the potential future direction of technology and our role in shaping it\"}),/*#__PURE__*/e(\"ol\",{start:\"14\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Artificial-Intelligence-Basics-Non-Technical-Introduction/dp/1484250273\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Artificial Intelligence Basics: A Non-Technical Introduction\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Tom Taulli\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/33dd024e-ff2e-4db8-af95-ab3c6dd99a8b\"}),/*#__PURE__*/e(\"p\",{children:\"Artificial Intelligence Basics is written to provide readers with a foundational understanding of AI and its significant impact. The book offers a clear, accessible introduction to key AI concepts like machine learning, deep learning, natural language processing, robotics, and more, making it suitable for those without a technical background who want to grasp the essentials of AI and its influence on various industries\u200B\"}),/*#__PURE__*/e(\"ol\",{start:\"15\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Life-3-0-Being-Artificial-Intelligence/dp/1101970316/ref=pd_lpo_sccl_1/144-0250447-2632956?pd_rd_w=jNcxV&content-id=amzn1.sym.116f529c-aa4d-4763-b2b6-4d614ec7dc00&pf_rd_p=116f529c-aa4d-4763-b2b6-4d614ec7dc00&pf_rd_r=ZN37ECZFMESP52ARXZ9F&pd_rd_wg=lTBkR&pd_rd_r=dc45a88b-0a58-4ea8-9d3b-7498bd11616f&pd_rd_i=1101970316&psc=1\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Life 3.0 Being Human in the Age of Artificial Intelligence\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Max Tegmark\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/e2327d0c-6d5f-416b-a2a1-86d04813ee5b\"}),/*#__PURE__*/e(\"p\",{children:\"MIT professor Max Tegmark offers an insightful exploration of AI's future impact on life on Earth and possibly beyond. Tegmark outlines different stages of life, from biological to cultural, and envisages a new era (Life 3.0) marked by advanced technology and artificial general intelligence (AGI) with self-redesign capabilities. He discusses the short-term effects of AI, such as technological unemployment and AI weaponry, and presents various potential futures, from beneficial AI to dystopian outcomes.\"}),/*#__PURE__*/e(\"ol\",{start:\"16\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Age-AI-Our-Human-Future/dp/0316273996/ref=pd_lpo_sccl_3/144-0250447-2632956?pd_rd_w=jNcxV&content-id=amzn1.sym.116f529c-aa4d-4763-b2b6-4d614ec7dc00&pf_rd_p=116f529c-aa4d-4763-b2b6-4d614ec7dc00&pf_rd_r=ZN37ECZFMESP52ARXZ9F&pd_rd_wg=lTBkR&pd_rd_r=dc45a88b-0a58-4ea8-9d3b-7498bd11616f&pd_rd_i=0316273996&psc=1\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"The Age of AI: And Our Human Future\\xa0\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Henry Kissinger, Eric Schmidt, Daniel Huttenlocher\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/ac05c895-20f6-4dda-9c65-57529b76d708\"}),/*#__PURE__*/e(\"p\",{children:\"Penned by prominent figures Henry Kissinger, Eric Schmidt, and Daniel Huttenlocher, dives into artificial intelligence's transformative impact on society and the existential questions it poses for the human condition. The book forecasts AI's far-reaching effects on various sectors, from science to warfare, and warns of the dangers of unchecked AI, like the creation of convincing falsehoods and AI-driven conflicts.\"}),/*#__PURE__*/i(\"p\",{children:[\"17. \",/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Power-Prediction-Disruptive-Artificial-Intelligence/dp/1647824192\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Power and Prediction: The Disruptive Economics of Artificial Intelligence\"})})]}),/*#__PURE__*/e(\"p\",{children:\"By Ajay Agrawal, Joshua Gans, Avi Goldfarb\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/d3d32e86-0fa7-412f-8b75-a0be030cae67\"}),/*#__PURE__*/e(\"p\",{children:\"This book is considered essential for economists, entrepreneurs, social scientists, and policy analysts who seek to understand and maximize AI's benefits while navigating its disruptive impact\u200B. It explores the role of AI as a pivotal technology in decision-making, where prediction and judgment are central. The authors, notable for their previous work \\\"Prediction Machines,\\\" delve into how AI's evolution is transferring the task of prediction from humans to machines, enhancing the speed and accuracy of decisions across various industries like finance, healthcare, and retail\u200B\"}),/*#__PURE__*/e(\"ol\",{start:\"18\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Brief-History-Artificial-Intelligence-Where/dp/1250770742\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"A Brief History of Artificial Intelligence: What It Is, Where We Are, and Where We Are Going\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Michael Wooldridge\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/bd939b3b-2fca-4aee-9feb-9eeaf2e7ad48\"}),/*#__PURE__*/e(\"p\",{children:\"An accessible journey through the evolution and potential future of AI. Wooldridge, a seasoned AI researcher, offers a balanced view, acknowledging the groundbreaking advances in AI, such as driverless cars and translation tools, while maintaining a realistic outlook on the aspirations and challenges that lie ahead. His book is an engaging and comprehensive guide that covers AI's history, its current state, and the transformative impact it may have on our future\u200B\"}),/*#__PURE__*/e(\"ol\",{start:\"19\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Human-Compatible-Artificial-Intelligence-Problem-ebook/dp/B07N5J5FTS\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Human Compatible: Artificial Intelligence and the Problem of Control\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Stuart Russell\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",src:\"blob:https://framer.com/506e5008-68c1-40ba-9a3f-fd5bd1117e70\"}),/*#__PURE__*/e(\"p\",{children:\"A profound exploration of the existential risks posed by advanced AI. It critiques the current AI model's focus on achieving human-specified goals without considering broader human values, which could lead to catastrophic outcomes. Russell proposes a new AI framework based on three principles that prioritize the realization and understanding of human preferences through machine learning from human behavior. His work has been lauded for its intellectual rigor and accessible style, making it a critical read for those interested in the future of AI and its ethical implications\u200B\"}),/*#__PURE__*/e(\"ol\",{start:\"20\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(r,{href:\"https://www.amazon.com/Scary-Smart-Future-Artificial-Intelligence/dp/1529077621\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Scary Smart: The Future of Artificial Intelligence and How You Can Save Our World\\xa0\"})})})})}),/*#__PURE__*/e(\"p\",{children:\"By Mo Gawdat\"}),/*#__PURE__*/e(\"p\",{children:\"Mo Gawdat presents a critical examination of AI's trajectory, offering a blueprint for harmonizing our future with advanced AI. Gawdat, the former Chief Business Officer at Google and bestselling author, delves into how AI, projected to be a billion times more intelligent than humans by 2049, can preserve our species rather than endanger it.\"})]}),[u]:\"\",[f]:!0,[w]:\"e56k2noiO\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"myPYYcMde\",[c]:\"The Hot Girl Singularity\",[h]:\"the-hot-girl-singularity\",[d]:\"2023-11-01T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/tWBGUuJIVFsSnfh0frwLLQSxNKU.png\",srcSet:\"https://framerusercontent.com/images/tWBGUuJIVFsSnfh0frwLLQSxNKU.png?scale-down-to=512 512w, https://framerusercontent.com/images/tWBGUuJIVFsSnfh0frwLLQSxNKU.png 1024w\"},\"\"),[p]:\"The Hot Girl Singularity\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(\"em\",{children:\"This article was originally published in Black Box, the author's newsletter. Subscribe \"}),/*#__PURE__*/e(r,{href:\"https://jwang18.substack.com/\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:/*#__PURE__*/e(\"em\",{children:\"here\"})})}),/*#__PURE__*/e(\"em\",{children:\". \"}),/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"That a generative model\u2019s performance deteriorates as it is trained on more generated data is something of an AI folk theorem. I first came across it in the now-canonical \",/*#__PURE__*/e(\"em\",{children:\"The New Yorker\"}),\" \",/*#__PURE__*/e(r,{href:\"https://www.newyorker.com/tech/annals-of-technology/chatgpt-is-a-blurry-jpeg-of-the-web\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"piece\"})}),\", but those who aren\u2019t familiar may find my recent encounter more entertaining. I was spending the day at an AI gaming startup. \u201CYou know,\u201D one of the co-founders said casually while unwrapping her lunch, \u201Cwe\u2019re headed towards a hot girl singularity.\u201D I had to ask.\"]}),/*#__PURE__*/e(\"p\",{children:\"It turns out the team was having trouble generating female characters with realistic proportions despite using several negative prompts. Their leading hypothesis was the internet is so full of hot girl pictures that the model had no idea what realistic means. \u201CAnd it gets worse,\u201D she continued. \u201CAll of the hot girls being generated right now means future models will be trained on even more hot girls. A couple models later, the internet will be nothing but pictures of hot girls.\u201D\"}),/*#__PURE__*/i(\"p\",{children:[\"As silly as this sounds, the intuition is right! Researchers just \",/*#__PURE__*/e(r,{href:\"https://arxiv.org/pdf/2305.17493v2.pdf\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"demonstrated\"})}),\" a generalization of this process, which they term \",/*#__PURE__*/e(\"em\",{children:\"model collapse. \"}),\"Their work has lots of interesting implications (I recommend Rob Horning\u2019s \",/*#__PURE__*/e(r,{href:\"https://robhorning.substack.com/p/probable-events-poison-reality\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"take\"})}),\") and I immediately thought of the ongoing Reddit blackout. Like many, I don\u2019t buy that Reddit is charging for API access solely to capture some of the value its content has created as training data. But it\u2019s also not an invalid point. Only last week, I wrote about how LLMs is making every platform into a source of \",/*#__PURE__*/e(r,{href:\"https://jwang18.substack.com/p/recruiting-platform\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"hiring signal\"})}),\". This is an understatement; LLMs actually make platforms more valuable for \",/*#__PURE__*/e(\"em\",{children:\"everything\"}),\".\\xa0\"]}),/*#__PURE__*/e(\"p\",{children:\"If we take this motive at face value, then the question is how can platforms monetize their content without gating access. Both users and the platforms themselves should want the platforms to be as open as possible: users need services that use APIs and platforms need users for content. The problem is that content is basically a public good and LLMs are free riders. Privatizing is one solution, but it\u2019s like treating a fever by starving the patient to death. Instead, Reddit should take inspiration from hot girls.\"}),/*#__PURE__*/i(\"p\",{children:[\"In their paper, the researchers explicitly call out model collapse as separate from data poisoning. In particular, model collapse is an inherent property of models recursively trained on generated data, whereas data poisoning is the result of manipulated training data. This means platforms don\u2019t need to introduce possibly unnatural engineered content to shake off free-riding models\u200A\u2014\u200Amodels trained on too much generated data will \",/*#__PURE__*/e(\"em\",{children:\"poison themselves\"}),\". All the platforms have to do is \",/*#__PURE__*/e(\"em\",{children:\"add naturalistic generated content to accelerate model collapse\"}),\". They can then sell the human version to LLM companies, for whom provenance matters, while keeping the full version open to all.\\xa0\"]}),/*#__PURE__*/i(\"p\",{children:[\"This is an elegant solution on multiple fronts. Ethically, accelerating model collapse is \",/*#__PURE__*/e(\"em\",{children:\"strictly defensive\"}),\"; an LLM is impacted only if it avoids paying for original content. Economically, it effectively turns platform content into a quasi-public good by \",/*#__PURE__*/e(\"em\",{children:\"raising excludability\"}),\". Functionally, it enables platforms to have their cake and eat it too. And mechanically, it harnesses an existing phenomenon instead of creating a new system. Platforms simply \",/*#__PURE__*/e(\"em\",{children:\"encourage\"}),\" successive models to exaggerate \",/*#__PURE__*/e(\"em\",{children:\"truly \"}),\"probable events into singularities (or more precisely, an \",/*#__PURE__*/e(r,{href:\"https://en.wikipedia.org/wiki/Attractor\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"attractor\"})}),\").\\xa0\"]}),/*#__PURE__*/e(\"p\",{children:\"It also seems to be a glaring breach of user trust. Users want to engage with other real users, which is why platforms spend considerable effort fighting fake accounts and bots. But I think it\u2019s possible to extend generated content without losing users. The easiest way is to fight it less, or rather, throttle the effort so the ratio is just human enough for users. Another way is setting up \u201Cghost towns\u201D that mimic real user content without real users. On the other hand, platforms can incentivize users to contribute more original content by having them pay a nominal amount to verify their posts in exchange for a share of model profits.\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"So Mr. Huffman, tear down this paywall! Everything on Reddit has already been scraped, so only future content matters anyway. End the blackout by opening up your API again. Focus on including generated content in ways that are palatable to redditors. Believe in the power of hot girls and model collapse, and the monetization shall follow. \u220E\"}),/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(\"em\",{children:\"Am I crazy or could this work? Let me know at \"}),/*#__PURE__*/e(r,{href:\"https://twitter.com/j_wang18\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:/*#__PURE__*/e(\"em\",{children:\"@jwang_18\"})})}),/*#__PURE__*/e(\"em\",{children:\" or reach out on \"}),/*#__PURE__*/e(r,{href:\"https://www.linkedin.com/in/jwang18/\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:/*#__PURE__*/e(\"em\",{children:\"LinkedIn\"})})}),/*#__PURE__*/e(\"em\",{children:\".\"})]})]}),[u]:\"Jiarui Wang\",[f]:!0,[w]:\"I8_efFBQ0\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"MmeqzJsbs\"},{id:\"wQ5TFlv2n\",[c]:\"China's Intelligentized Warfare\",[h]:\"china-s-intelligentized-warfare\",[d]:\"2023-10-30T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/j2GV3xmOXDWlpES1BBM9oukzg.png\",srcSet:\"https://framerusercontent.com/images/j2GV3xmOXDWlpES1BBM9oukzg.png?scale-down-to=512 512w, https://framerusercontent.com/images/j2GV3xmOXDWlpES1BBM9oukzg.png 1024w\"},\"\"),[p]:\"China's Intelligentized Warfare\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/i(\"p\",{children:[\"The Department of Defense (DoD) just released their \",/*#__PURE__*/e(r,{href:\"https://media.defense.gov/2023/Oct/19/2003323409/-1/-1/1/2023-MILITARY-AND-SECURITY-DEVELOPMENTS-INVOLVING-THE-PEOPLES-REPUBLIC-OF-CHINA.PDF\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"annual report\"})}),\" on military developments in the China, calling it the only competitor to the US with the intent and capacity to reshape international order.\"]}),/*#__PURE__*/e(\"p\",{children:\"DoD requests for bilateral defense engagements and communications to their Chinese counterparts were denied at multiple levels - showing an unwillingness to collaborate by China.\"}),/*#__PURE__*/e(\"p\",{children:\"Technology and AI was front and center in the new report. China is rapidly expanding their usage of AI at every level of warfare, as they see future conflicts will primarily be fought with information in what they call \u201Cintelligentized warfare\u201D.\"}),/*#__PURE__*/e(\"p\",{children:\"The report warns of a new concept being employed by China\u2019s People\u2019s Liberation Army (PLA) known as \u201CMulti-Domain Precision Warfare\u201D (MDPW) that aims to leverage AI and big data to quickly identify vulnerabilities in US systems and use \u201Cmultiple domains\u201D to launch precision strikes directly on those vulnerabilities. For example, the ability to recognize a cyber vulnerability in a natural gas pipeline in the US and knock it offline for potentially weeks at a time.\"}),/*#__PURE__*/i(\"p\",{children:[\"The Chinese military is heavily investing in domestic hardware companies, especially the semiconductor chip industry, to reduce reliance on foreign suppliers which they see as one of their biggest vulnerabilities. China has also invested over $1B to build a national quantum computing lab, the biggest investment in quantum by a government to date, to help further their technology capabilities. \",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"While the Chinese government has long been known to closely monitor and collaborate with leaders in their tech industry, China is actively working to increase the proportion of private companies contributing to military projects. The DoD report outlined key formal initiatives between China\u2019s military and domestic private companies that promotes R&D collaboration and domain resource sharing. This means outside (US) investments in China\u2019s tech sector will almost certainly be helping to fund new technologies that will be weaponized by the Chinese military. \",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"With the latest DoD report, it's clear that China is intensively focused on AI to fortify its military prowess and position itself as the world\u2019s leading military force. The increasing collaboration between China's private tech sector and its military raises concerns for international stakeholders, especially for the U.S., which needs to be cautious about its technological investments in China.\"]})]}),[u]:\"\",[f]:!0,[w]:\"CTvKjSNWG\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"GhXKhAwI3\"},{id:\"qTu15HYoA\",[c]:\"China's Digital Humans Displacing Workers\",[h]:\"china-s-digital-humans\",[d]:\"2023-10-16T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/jxOvgjucDi7VaWcYQSS3CGaCEAY.png\",srcSet:\"https://framerusercontent.com/images/jxOvgjucDi7VaWcYQSS3CGaCEAY.png?scale-down-to=512 512w, https://framerusercontent.com/images/jxOvgjucDi7VaWcYQSS3CGaCEAY.png 1024w\"},\"\"),[p]:\"China's Digital Humans Displacing Workers\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/i(\"p\",{children:[\"AI-generated avatars, or \u201Cdigital humans\u201D are becoming increasingly popular in China. \",/*#__PURE__*/e(r,{href:\"https://www.chinadaily.com.cn/a/202209/21/WS632a70baa310fd2b29e78ded.html\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Beijing officials have estimated\"})}),\" that the digital human sector will surpass by $50 billion yuan (~$7B USD) by 2025 and $38.5B by 2030, according to a report released by QbitAI. There are over 280,000 Chinese companies actively working in the digital human space. These digital humans exist solely on the internet and are typically divided into two categories: character-based and function-based.\",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"strong\",{children:\"Character-based\"}),\" digital humans are often created in the likeness of celebrities or IP from pop culture characters.\",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"strong\",{children:\"Function-based\"}),\" digital humans, are often used for their reduced cost and efficiency compared to using real people, and can take the form of musicians, fashion models, news anchors, influencers and even financial advisors.\",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"These function-based digital humans are putting some people out of work, especially in the live commerce sector. Live commerce, similar to QVC in America, is incredibly popular in China. There are over 1 million people working as live commerce hosts in China, having brought in over \",/*#__PURE__*/e(r,{href:\"https://www.sixthtone.com/news/1013597\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"$140B in goods sold last year\"})}),\". Chinese tech giants like Tencent and Baidu (the Google of China) both offer platforms that allow for the creation of AI-generated avatars that can livestream 24/7 at a much cheaper cost than paying a real person to host. Baidu\u2019s digital avatar platform \",/*#__PURE__*/e(r,{href:\"https://www.synthesia.io/\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"XiLing\"})}),\", which is comparable to US-based \",/*#__PURE__*/e(r,{href:\"https://www.synthesia.io/\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Synthesia\"})}),\", allows for creation of digital avatars from existing IP or custom creation based on industry and allows for translation to multiple languages.\"]}),/*#__PURE__*/e(o.div,{className:\"framer-text-module\",style:{\"--aspect-ratio\":\"560 / 315\",aspectRatio:\"var(--aspect-ratio)\",height:\"auto\",width:\"100%\"},children:/*#__PURE__*/e(n,{componentIdentifier:\"module:NEd4VmDdsxM3StIUbddO/bZxrMUxBPAhoXlARkK9C/YouTube.js:Youtube\",children:i=>/*#__PURE__*/e(l,{...i,play:\"Off\",shouldMute:!0,thumbnail:\"High Quality\",url:\"https://www.youtube.com/watch?v=lvMS4nHFBSI\"})})}),/*#__PURE__*/e(\"p\",{children:\"The rise of AI-generated avatars, particularly in China, underscores a significant shift in the digital landscape. While they offer cost efficiency and scalability, particularly for companies like Tencent and Baidu, there's an undeniable impact on employment in certain sectors. The staggering projections by Beijing officials highlight not only the commercial potential but also the broad cultural adoption expected in the coming years. As hundreds of thousands of companies dive into the digital human space, one can't help but wonder how this will reshape industries, societal norms, and individual careers in the not-so-distant future. As we stand on the cusp of this AI evolution, it signals us to adapt, innovate, and reconsider the essence of human roles in an increasingly digital world.\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"br\",{className:\"trailing-break\"})}),/*#__PURE__*/i(\"h4\",{children:[\"Originally featured in \",/*#__PURE__*/e(r,{href:\"https://amidai.substack.com/p/the-ai-china-report-4\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"The AI China Report #4\"})})]})]}),[u]:\"\",[f]:!1,[w]:\"CTvKjSNWG\",[y]:!1,[b]:\"Z8WWlniBM\",[v]:\"GhXKhAwI3\"},{id:\"OS_5HEkos\",[c]:\"Governments Using AI for Deepfakes\",[h]:\"governments-using-ai-for-deepfakes\",[d]:\"2023-10-09T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/wHyuMaM9bsYW3B1SVK5tRSkXg.png\",srcSet:\"https://framerusercontent.com/images/wHyuMaM9bsYW3B1SVK5tRSkXg.png?scale-down-to=512 512w, https://framerusercontent.com/images/wHyuMaM9bsYW3B1SVK5tRSkXg.png 1024w\"},\"\"),[p]:\"Governments Using AI for Deepfakes\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/i(\"p\",{children:[\"A \",/*#__PURE__*/e(r,{href:\"https://freedomhouse.org/report/freedom-net/2023/repressive-power-artificial-intelligence\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"new report from Freedom House\"})}),\", a D.C.-based nonprofit political advocacy group, found that at least 16 countries over the last year have used deepfake technology to \u201Csow doubt, smear opponents, or influence public debate.\u201D\"]}),/*#__PURE__*/i(\"p\",{children:[\"Earlier this year, during national protests in Pakistan, Former Prime Minister Imran Khan (who is now imprisoned) posted a \",/*#__PURE__*/e(r,{href:\"https://twitter.com/ImranKhanPTI/status/1659621414473986048?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E1659621414473986048%7Ctwgr%5E1125e880db32a38dccf6bb72686494376b32457a%7Ctwcon%5Es1_&ref_url=https%3A%2F%2Fgizmodo.com%2Fembed%2Finset%2Fiframe%3Fid%3Dtwitter-1659621414473986048autosize%3D1\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"video on X\"})}),\" to rally his supporters. That video used deepfake technology to show an image of a protestor standing in front of riot police - this image was generated by AI.\"]}),/*#__PURE__*/i(\"p\",{children:[\"Former President Donald Trump and Florida Governor Ron DeSantis have both used deepfake technology to attack each other in political ads as the Republican primary competition has heated up this year. Trump ran a \",/*#__PURE__*/e(r,{href:\"https://gizmodo.com/desantis-trump-musk-twitter-hitler-deepfake-campaign-1850475254\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"fake Twitter Spaces event\"})}),\" that showed DeSantis in a conversation with George Soros, Adolph Hitler, and\u2026. The Devil. The DeSantis team fought back using \",/*#__PURE__*/e(r,{href:\"https://www.nytimes.com/2023/06/08/us/politics/desantis-deepfakes-trump-fauci.html\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"AI to create fake images\"})}),\" of Trump hugging and kissing Dr. Anthony Fauci.\"]}),/*#__PURE__*/i(\"p\",{children:[\"China has been hiring private companies to deploy AI-generated propaganda that mimics newscasters, \",/*#__PURE__*/e(r,{href:\"https://www.nytimes.com/2023/02/07/technology/artificial-intelligence-training-deepfake.html\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"they even created an AI-generated news division called \u201CWolf News\u201D\"})}),\" that shares video clips of human characters presented as newscasters. \",/*#__PURE__*/e(r,{href:\"https://www.vice.com/en/article/z34jge/venezuela-ai-newscaster-disinformation\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Venezuela has employed similar tactics\"})}),\" using U.S.-based AI avatar company \",/*#__PURE__*/e(r,{href:\"https://www.synthesia.io/?utm_term=synthesia&utm_campaign=Synthesia&utm_source=google&utm_medium=cpc&hsa_acc=5132031546&hsa_cam=16607623891&hsa_grp=135039912935&hsa_ad=665286628726&hsa_src=g&hsa_tgt=kwd-323407513834&hsa_kw=synthesia&hsa_mt=e&hsa_net=adwords&hsa_ver=3&gclid=CjwKCAjwg4SpBhAKEiwAdyLwvOok64U-RZBXvw-bERyMs_ZWVLuRZyZSZU2udFPZs_iRkq5lYU7_HxoCwK8QAvD_BwE\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Synthesia\"})}),\".\"]}),/*#__PURE__*/e(\"p\",{children:\"All of this has an important consequence: people no longer know what is real and what is deepfaked. The lines of reality are becoming blurred. This allows political actors to call out video and audio recordings as deepfakes, even if they are genuine, further spreading doubt over all content. The technology is leading us to a world where you can no longer trust what you see, read, or hear on the internet as genuine content - even if it looks or sounds like a real well-known political figure.\"}),/*#__PURE__*/e(\"p\",{children:\"Aside from using AI to create false narratives, it was found that at least 22 countries had processes in place for social media companies to use AI to identify and remove dissident political speech from the internet. The report found that a record 41 governments, including the U.S. and European countries, blocked websites with content that should be protected under free expression standards within international human rights law.\"}),/*#__PURE__*/e(\"p\",{children:\"Deepfakes will only get better as AI technology advances. Without proper laws and regulations, it will become harder to decipher what is real and what is fake on the internet. How we handle this issue will have lasting implications for humans everywhere.\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"br\",{className:\"trailing-break\"})}),/*#__PURE__*/i(\"h4\",{children:[\"Originally featured in \",/*#__PURE__*/e(r,{href:\"https://amidai.substack.com/p/the-ai-china-report-3\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"The AI China Report #3\"})})]})]}),[u]:\"\",[f]:!1,[w]:\"CTvKjSNWG\",[y]:!1,[b]:\"B5rd62fyE\",[v]:\"GhXKhAwI3\"},{id:\"DihDm_Iqy\",[c]:\"China's New AI Rules Could Affect U.S. Companies\",[h]:\"china-s-new-ai-rules-could-affect-u-s-companies\",[d]:\"2023-10-02T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/eMf6yQHGVxlR06gXcj9TnaECYo.png\",srcSet:\"https://framerusercontent.com/images/eMf6yQHGVxlR06gXcj9TnaECYo.png?scale-down-to=512 512w, https://framerusercontent.com/images/eMf6yQHGVxlR06gXcj9TnaECYo.png 1024w\"},\"robot looking over the shoulder of chinese regulators\"),[p]:\"China's New AI Rules Could Affect U.S. Companies\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Last month the The Cyberspace Administration of China (CAC) put into effect new rules for generative AI tools, which require AI companies to register with the government and pass a security review before they can be released to the public.\"}),/*#__PURE__*/e(\"p\",{children:\"The rules also require that all data used for training AI models come from \u201Clegitimate sources\u201D and don\u2019t infringe on national IP rights. Companies creating AI tools can also be held responsible for the content that is created on their platforms including content that advocates for terrorism, extremism, ethnic hatred, violence, obscenity, and fake or harmful information.\"}),/*#__PURE__*/e(\"p\",{children:\"These rules create exposure for AI companies, as it has proven difficult to prevent generative AI tools (especially chatbots like ChatGPT) from creating content that could violate these rules.\"}),/*#__PURE__*/i(\"p\",{children:[\"U.S. technology companies are still ahead of China in terms of AI capability, with China\u2019s best models being comparable to OpenAI\u2019s GPT-3 (which is a two year old model, OpenAI has since released GPT-3.5 and GPT-4 to the public). It\u2019s likely that these regulations could make it more difficult for Chinese companies to innovate at a speed that will allow them to catch up the the U.S.\",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"The U.S. still doesn\u2019t have AI regulations, but is actively working on putting together guidelines as top technology leaders like Elon Musk, Mark Zuckerberg, and Sundar Pichai of Google, among others, \",/*#__PURE__*/e(r,{href:\"https://www.nytimes.com/2023/09/13/technology/silicon-valley-ai-washington-schumer.html\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"met in a closed door session with U.S. lawmakers\"})}),\" earlier this month to discuss regulation.\"]}),/*#__PURE__*/e(\"p\",{children:\"One of the most important questions regarding U.S. regulation is how to strike a balance that allows for protection of Americans (and jobs) without hindering AI innovation to the point that the U.S. loses their lead to China or other countries.\"}),/*#__PURE__*/i(\"p\",{children:[\"Lawmakers will be looking closely at China and the CAC rules as they decide on the best approach for AI regulation in the U.S. The severe regulations in China could prompt the U.S. to also create stricter guidelines for its own AI companies. This would affect new AI startups and especially current tech titans like Google, Meta, Amazon, and Microsoft that are already heavily investing in and developing AI products.\",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"Stronger U.S. regulation could lead to greater protections for Americans, but hinder companies that have been improving AI technologies at a rapid pace. However, this could introduce the risk of falling behind if the Chinese government chooses not to enforce their own regulations on certain national companies that it favors.\"]}),/*#__PURE__*/e(\"h4\",{children:/*#__PURE__*/e(\"br\",{className:\"trailing-break\"})}),/*#__PURE__*/i(\"h4\",{children:[\"Originally featured in \",/*#__PURE__*/e(r,{href:\"https://amidai.substack.com/p/the-ai-china-report-2\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"The AI China Report #2\"})})]})]}),[u]:\"\",[f]:!1,[w]:\"CTvKjSNWG\",[y]:!1,[b]:\"B5rd62fyE\",[v]:\"GhXKhAwI3\"},{id:\"nkETB09b4\",[c]:\"Huawei's Secret Chip Factories\",[h]:\"huawei-s-secret-chip-factories\",[d]:\"2023-09-25T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/UbRihPyrm74eVWOj0Kqu3AhJ7Gs.jpg\",srcSet:\"https://framerusercontent.com/images/UbRihPyrm74eVWOj0Kqu3AhJ7Gs.jpg?scale-down-to=512 512w, https://framerusercontent.com/images/UbRihPyrm74eVWOj0Kqu3AhJ7Gs.jpg?scale-down-to=1024 1024w, https://framerusercontent.com/images/UbRihPyrm74eVWOj0Kqu3AhJ7Gs.jpg 1200w\"},\"\"),[p]:\"Huawei's Secret Chip Factories\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Tensions between the U.S. and China have continued to heat up over the past year, especially after the mainstream interest and boom of adoption in AI technology that shook the world last year.\"}),/*#__PURE__*/e(\"p\",{children:\"It began on October 7, 2022 when the Biden administration announced sweeping restrictions on the sale of semiconductor technology to China, which many considered \u201Ca declaration of economic war on China\u201D.\"}),/*#__PURE__*/e(\"p\",{children:\"The restrictions limited U.S. exports of semiconductor chips, which use graphic processing units (GPUs) to power AI applications. This included setting a cap on the maximum processing speed of chips that could be sold to China. The goal being to slow China\u2019s development of new AI technologies, as fears have grown of a winner takes all scenarios - both from an economic and military perspective - in the race to develop the most powerful AI tech.\"}),/*#__PURE__*/e(\"p\",{children:\"Enter Huawei Technologies, one of China\u2019s leading telecommunications companies and the second-biggest smartphone maker in the world, after Samsung. Last year, the Chinese tech giant received an estimated $30 billion in funding from the Chinese government to move into chip production.\"}),/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(r,{href:\"https://www.bloomberg.com/news/articles/2023-08-23/huawei-building-secret-chip-plants-in-china-to-bypass-us-sanctions-group-warns?srnd=all&sref=y3YMCJ4e\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/i(\"a\",{children:[/*#__PURE__*/e(\"em\",{children:\"Bloomberg\"}),\" recently reported\"]})}),\" that Huawei is building a secret string of semiconductor factories across China to get around the U.S. sanctions and make China less reliant on chips from the U.S. and Taiwan - the two current leaders in advanced chip production. Huawei itself was put on a blacklist by the Biden administration in January which blocks U.S. companies from selling and exporting to Huawei.\"]}),/*#__PURE__*/e(\"p\",{children:\"U.S.-based Nvidia (NASDAQ: NVDA) is one of the global leading chip makers, having recently experienced a doubling of its stock price due to demand for its GPU chips to create and power new AI technology. For context, Nvidia is now the 6th largest public company in the U.S. joining the likes of Apple, Amazon, and Microsoft as one of the few companies with a $1 trillion market cap.\"}),/*#__PURE__*/i(\"p\",{children:[\"Nvidia responded to U.S. restrictions by creating new, less powerful chip products specifically for China that fall under the limit of processing speeds and data transfer rates imposed by the Biden administration. This led to Chinese \",/*#__PURE__*/e(r,{href:\"https://www.ft.com/content/9dfee156-4870-4ca4-b67d-bb5a285d855c?utm_source=stevens-newsletter-009afe.beehiiv.com&utm_medium=newsletter&utm_campaign=the-ai-china-brief-issue-1\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"tech giants placing orders\"})}),\" for over $5 billion worth of these handicapped chips. The U.S. is now considering increasing restrictions that would stop all shipments of chips made by Nvidia and other U.S. makers to Chinese companies without first obtaining a license.\"]}),/*#__PURE__*/e(\"p\",{children:\"These less powerful chips will likely make it more difficult for Chinese tech companies to train powerful AI models than their U.S. counterparts, giving U.S. companies a head start in developing stronger models. The rate of improvement of AI tech has been growing incredibly fast, and the U.S. hopes that this head start will allow them to widen the gap over time to the point where China cannot catch up, even when they develop the capabilities to make comparable chips themselves.\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"br\",{className:\"trailing-break\"})}),/*#__PURE__*/i(\"h4\",{children:[\"Originally featured in \",/*#__PURE__*/e(r,{href:\"https://amidai.substack.com/p/the-ai-china-report-1\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"The AI China Report #1\"})})]})]}),[u]:\"\",[f]:!1,[w]:\"CTvKjSNWG\",[y]:!0,[b]:\"B5rd62fyE\",[v]:\"GhXKhAwI3\"},{id:\"KSb0WOa6e\",[c]:\"AI + Education: The Unprecedented Impact of AI like ChatGPT on Education\",[h]:\"ai-education-the-unprecedented-impact-of-ai-like-chatgpt-on-education-in-the-united-states\",[d]:\"2023-07-21T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/vCTN1U8RPOkBI9Nhl1TzRp09VV0.jpg?scale-down-to=4096\",srcSet:\"https://framerusercontent.com/images/vCTN1U8RPOkBI9Nhl1TzRp09VV0.jpg?scale-down-to=512 512w, https://framerusercontent.com/images/vCTN1U8RPOkBI9Nhl1TzRp09VV0.jpg?scale-down-to=1024 1024w, https://framerusercontent.com/images/vCTN1U8RPOkBI9Nhl1TzRp09VV0.jpg?scale-down-to=2048 2048w, https://framerusercontent.com/images/vCTN1U8RPOkBI9Nhl1TzRp09VV0.jpg?scale-down-to=4096 4096w, https://framerusercontent.com/images/vCTN1U8RPOkBI9Nhl1TzRp09VV0.jpg 6000w\"},\"\"),[p]:\"\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Education, the foundation of personal and societal growth, is on the brink of an unprecedented revolution. The catalyst? The explosion of artificial intelligence (AI) technologies such as OpenAI's groundbreaking language model, ChatGPT. This revolution promises a future where individual learning needs are met with precision, education becomes universally accessible, and the role of teachers evolves to harness and foster human potential.\"}),/*#__PURE__*/e(\"h3\",{children:/*#__PURE__*/e(\"strong\",{children:\"The Emergence of AI in Education\"})}),/*#__PURE__*/e(\"p\",{children:\"AI technologies like ChatGPT utilize an advanced technology known as natural language processing (NLP). Trained on extensive volumes of text data, these AI models have the capability to generate human-like text, participate in intelligent conversations, answer queries, and create content. Such tools are not just impressive; they're revolutionizing the way we teach and learn (Belpaeme et al., 2018).\"}),/*#__PURE__*/e(\"h3\",{children:/*#__PURE__*/e(\"strong\",{children:\"The Promise of Personalized Learning\"})}),/*#__PURE__*/e(\"p\",{children:\"Each student is unique in their learning style, pace, and needs. However, one-on-one tutoring, the ideal approach to personalized learning, is often a luxury in the conventional U.S. educational landscape due to resource constraints. AI tutors like ChatGPT can fill this gap. These intelligent systems adapt to individual learning styles, track progress, and provide personalized guidance, thereby promoting a mastery-based learning environment (Woolf et al., 2013).\"}),/*#__PURE__*/e(\"h3\",{children:/*#__PURE__*/e(\"strong\",{children:\"Making Education Universally Accessible\"})}),/*#__PURE__*/e(\"p\",{children:\"Geographical location, economic disparity, or lack of quality teachers should not obstruct the path to quality education. The beauty of AI-driven education tools like ChatGPT lies in their potential to democratize learning. All a student needs is an internet connection, and they have a personal tutor at their fingertips. This could drastically reshape the education landscape in under-resourced areas and could be a significant step toward educational equity.\"}),/*#__PURE__*/e(\"h3\",{children:/*#__PURE__*/e(\"strong\",{children:\"Evolving the Role of Educators\"})}),/*#__PURE__*/e(\"p\",{children:\"AI is not a replacement for human educators; instead, it\u2019s an invaluable tool that enhances their capabilities. AI systems can manage administrative tasks, grading, and basic content delivery, freeing educators to invest their time and energy in nurturing critical thinking, creativity, and socio-emotional learning in students \u2013 skills crucial for success in the 21st century (Bughin & Hazan, 2017).\"}),/*#__PURE__*/e(\"h3\",{children:/*#__PURE__*/e(\"strong\",{children:\"Navigating the Challenges\"})}),/*#__PURE__*/e(\"p\",{children:\"Though promising, the incorporation of AI technologies like ChatGPT into education poses several challenges. The digital divide, or the gap in access to technology, remains a pressing issue in the U.S. AI adoption must coincide with comprehensive measures to bridge this divide and ensure no student is left behind.\"}),/*#__PURE__*/e(\"p\",{children:\"Data privacy is another significant concern. As AI models learn from data, the potential for misuse of sensitive student data is a real threat. It's essential to create stringent data privacy policies to safeguard student information (Zeide, 2017).\"}),/*#__PURE__*/e(\"p\",{children:\"Furthermore, while AI models like ChatGPT can simulate human conversation impressively well, they lack the ability to fully comprehend and respond to complex human emotions. The presence of human teachers, capable of empathy, remains irreplaceable.\"}),/*#__PURE__*/e(\"h3\",{children:/*#__PURE__*/e(\"strong\",{children:\"The Future is Now\"})}),/*#__PURE__*/e(\"p\",{children:\"With rapid advancements in technology, AI-driven learning tools like ChatGPT will become an integral part of the U.S. educational system. While the transition poses its challenges, the potential benefits \u2013 personalized learning experiences, universal access, and evolved teacher roles \u2013 make it a journey worth undertaking.\"}),/*#__PURE__*/e(\"p\",{children:\"In conclusion, AI technologies like ChatGPT are set to disrupt the U.S. educational landscape in ways we never thought possible. By personalizing learning, democratizing access, and reshaping the role of educators, we stand at the precipice of an educational renaissance. As we move forward, the focus should remain on how we can leverage AI to augment human potential, fostering an educational environment that is inclusive, engaging, and adaptable to the needs of every student.\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"br\",{className:\"trailing-break\"})}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"References:\"})}),/*#__PURE__*/i(\"ul\",{children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Belpaeme, T., Kennedy, J., Ramachandran, A., Scassellati, B., & Tanaka, F. (2018). Social robots for education: A review. Science Robotics, 3(21).\"})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Woolf, B. P., Lane, H. C., Chaudhri, V. K., & Kolodner, J. L. (2013). AI grand challenges for education. AI Magazine, 34(4), 9-26.\"})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(\"em\",{children:\"Bughin, J., & Hazan, E. (2017). \"}),/*#__PURE__*/e(r,{href:\"https://www.mckinsey.com/capabilities/quantumblack/our-insights/how-artificial-intelligence-can-deliver-real-value-to-companies\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:/*#__PURE__*/e(\"em\",{children:\"How artificial intelligence can deliver real value to companies\"})})}),/*#__PURE__*/e(\"em\",{children:\". McKinsey Global Institute. \"})]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Zeide, E. (2017). The structural consequences of big data-driven education. Big Data, 5(2), 164-172.\"})})})]})]}),[u]:\"\",[f]:!0,[w]:\"ZUJZiBqb2\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"AWq5tY3Er\"},{id:\"Glv7twSJa\",[c]:\"10 Reasons Teachers and Educators must pay attention to ChatGPT and AI technologies\",[h]:\"10-reasons-teachers-and-educators-must-pay-attention-to-chatgpt-and-ai-technologies\",[d]:\"2023-07-19T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/y09zYgs7XW66xMvCnGVL3Na11M.jpg?scale-down-to=4096\",srcSet:\"https://framerusercontent.com/images/y09zYgs7XW66xMvCnGVL3Na11M.jpg?scale-down-to=512 512w, https://framerusercontent.com/images/y09zYgs7XW66xMvCnGVL3Na11M.jpg?scale-down-to=1024 1024w, https://framerusercontent.com/images/y09zYgs7XW66xMvCnGVL3Na11M.jpg?scale-down-to=2048 2048w, https://framerusercontent.com/images/y09zYgs7XW66xMvCnGVL3Na11M.jpg?scale-down-to=4096 4096w, https://framerusercontent.com/images/y09zYgs7XW66xMvCnGVL3Na11M.jpg 6000w\"},\"\"),[p]:\"\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"ol\",{children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h4\",children:/*#__PURE__*/e(\"h4\",{children:\"Harnessing AI for Continuous Learning: Facilitating 24/7 Education\"})})}),/*#__PURE__*/e(\"p\",{children:\"AI systems like ChatGPT function around the clock, promoting continuous and self-paced learning, which research shows can lead to better student engagement and outcomes (Baker, 2016).\"}),/*#__PURE__*/e(\"ol\",{start:\"2\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h4\",children:/*#__PURE__*/e(\"h4\",{children:\"Multilingual Support in AI: Bridging Linguistic Gaps in Classrooms\"})})}),/*#__PURE__*/e(\"p\",{children:\"AI technologies support multiple languages, thereby aiding ESL (English as a Second Language) students and promoting linguistic inclusivity in classrooms (Plass et al., 2020).\"}),/*#__PURE__*/e(\"ol\",{start:\"3\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h4\",children:/*#__PURE__*/e(\"h4\",{children:\"AI and Special Education: Tailored Learning for Special Needs Students\"})})}),/*#__PURE__*/e(\"p\",{children:\"Certain AI technologies can be tailored to accommodate students with learning disabilities, providing them with unique learning pathways (Luckin et al., 2016).\"}),/*#__PURE__*/e(\"ol\",{start:\"4\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h4\",children:/*#__PURE__*/e(\"h4\",{children:\"The Power of Real-Time Progress Monitoring in AI-Enabled Education\"})})}),/*#__PURE__*/e(\"p\",{children:\"AI tools can track student performance in real-time, offering actionable insights for educators to adapt their instruction (Spector, 2014).\"}),/*#__PURE__*/e(\"ol\",{start:\"5\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h4\",children:/*#__PURE__*/e(\"h4\",{children:\"Streamlining Education: AI in Content Creation and Curation\"})})}),/*#__PURE__*/e(\"p\",{children:\"AI models like ChatGPT can generate educational content, formulate engaging quizzes, or recommend resources tailored to student's needs, thereby saving educators' time and resources (Graesser et al., 2001).\"}),/*#__PURE__*/e(\"ol\",{start:\"6\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h4\",children:/*#__PURE__*/e(\"h4\",{children:\"Reducing the Administrative Burden\"})})}),/*#__PURE__*/e(\"p\",{children:\"AI can automate administrative tasks like scheduling, attendance tracking, or grading, allowing educators to devote more time to pedagogy (Bui et al., 2018).\"}),/*#__PURE__*/e(\"ol\",{start:\"7\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h4\",children:/*#__PURE__*/e(\"h4\",{children:\"Upgrading Teacher Skills: AI in Professional Development for Educators\"})})}),/*#__PURE__*/e(\"p\",{children:\"AI can be a tool for teachers' professional development, assisting them in staying updated with the latest educational research and trends (King, 2016).\"}),/*#__PURE__*/e(\"ol\",{start:\"8\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h4\",children:/*#__PURE__*/e(\"h4\",{children:\"Breaking Boundaries: Promoting Global Collaboration through AI in Education\"})})}),/*#__PURE__*/e(\"p\",{children:\" AI breaks down geographical and language barriers, fostering global collaboration among educators (Mishra et al., 2019).\"}),/*#__PURE__*/e(\"ol\",{start:\"9\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h4\",children:/*#__PURE__*/e(\"h4\",{children:\"Predictive Analysis in AI: Anticipating Student Outcomes for Proactive Teaching Strategies\"})})}),/*#__PURE__*/e(\"p\",{children:\"AI's predictive analysis can anticipate future outcomes such as student performance, providing proactive strategies for educators (Dietterich, 2000).\"}),/*#__PURE__*/e(\"ol\",{start:\"10\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h4\",children:/*#__PURE__*/e(\"h4\",{children:\"Enhancing Parent-Teacher Communication: The Role of AI in Fostering Transparency in Education\"})})}),/*#__PURE__*/e(\"p\",{children:\"AI can streamline communication between parents and teachers, enhancing transparency in the learning process (Epstein & Sheldon, 2002).\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"br\",{className:\"trailing-break\"})}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"References:\"})}),/*#__PURE__*/i(\"ul\",{children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Baker, R. (2016). Stupid Tutoring Systems, Intelligent Humans. International Journal of Artificial Intelligence in Education, 26(2), 600-614.\"})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Plass, J.L., Pawar, S., MacNamara, A., et al. (2020). Designing for positive language learning experiences in a multimodal, AI-driven learning environment. AI & Society, 35, 941\u2013954.\"})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Luckin, R., Holmes, W., Griffiths, M., & Forcier, L.B. (2016). Intelligence Unleashed: An argument for AI in Education. Pearson.\"})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Spector, J.M. (2014). Conceptualizing the emerging field of smart learning environments. Smart Learning Environments, 1(1), 2.\"})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Graesser, A.C., Chipman, P., Leeming, F., & Biedenbach, S. (2009). Deep learning and emotion in serious games. In Serious game design and development: Technologies for training and learning.\"})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Bui, H., Sylvestre, J., & Zaihrayeu, I. (2018). AI in education: the need for critical reflection. Ethics and Information Technology, 20(4), 279-285.\"})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"King, F. (2016). Teacher professional development to support the integration of ICT in teaching and learning: A systemic approach. Journal of Computer Assisted Learning, 32(4), 299-311.\"})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Mishra, P., Henriksen, D., Boltz, L.O., Richardson, C. (2019). Creativity, Technology, and Teacher Education. In Handbook of Research on TPACK in the Digital Age.\"})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Dietterich, T. (2000). An Experimental Comparison of Three Methods for Constructing Ensembles of Decision Trees: Bagging, Boosting, and Randomization. Machine Learning, 40(2), 139-157.\"})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Epstein, J.L., & Sheldon, S.B. (2002). Present and accounted for: Improving student attendance through family and community involvement. The Journal of Educational Research, 95(5), 308-318.\"})})})]})]}),[u]:\"\",[f]:!0,[w]:\"ZUJZiBqb2\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"AWq5tY3Er\"},{id:\"OQygOhWu0\",[c]:\"The Surprising Prescience of \u201CMinority Report\u201D\",[h]:\"the-surprising-prescience-of-minority-report\",[d]:\"2023-06-01T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/6bt3dYNjyHdnrDniP9DxIWmJPlQ.jpeg\",srcSet:\"https://framerusercontent.com/images/6bt3dYNjyHdnrDniP9DxIWmJPlQ.jpeg?scale-down-to=512 512w, https://framerusercontent.com/images/6bt3dYNjyHdnrDniP9DxIWmJPlQ.jpeg 856w\"},\"\"),[p]:\"\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"In 2002, Steven Spielberg released the science fiction film Minority Report, set in a futuristic Washington D.C. in the year 2054. The film depicted a world with widespread use of AI technologies, including autonomous vehicles, personalized advertising, and predictive policing. At the time, many of these concepts seemed far-fetched. Yet today, less than 20 years later, the world is quickly advancing to resemble the one Spielberg envisioned.\"}),/*#__PURE__*/e(\"p\",{children:\"The film centers around the \u201CPreCrime\u201D police department, which uses AI systems to predict and prevent murders. The system scans the minds of three psychics to generate names and images of soon-to-be murderers and their victims. Officers then conduct an \u201Carrest\u201D before any crime is actually committed. Setting aside the fantastical mind-scanning element, the notion of predictive policing using AI algorithms now seems eerily plausible. Police departments are increasingly turning to \u201Cheat list\u201D systems that use historical data to predict individuals likely to be involved in future crimes. While still controversial, these systems are being deployed in cities like Chicago and Los Angeles.\"}),/*#__PURE__*/e(\"p\",{children:\"Autonomous vehicles, depicted in the film using a maglev highway system, are now on the verge of becoming mainstream. Companies like Tesla, Uber, and Google are aggressively developing and testing self-driving vehicle technology. While we likely won\u2019t see roads specifically for autonomous vehicles, many experts predict they could make up the majority of vehicles on the road by 2030.\"}),/*#__PURE__*/e(\"p\",{children:\"In Minority Report, protagonist John Anderton is bombarded with personalized ads as he walks through public spaces. The ads scan his eyes to identify him and pitch products based on his interests and past purchases. Today, personalized and targeted advertising is ubiquitous on platforms like Facebook, Google, and Amazon. These companies track our interactions, searches, and purchases to build detailed profiles of our interests and behaviors. They then use these profiles to tailor ads and product recommendations. While retail stores can\u2019t yet scan our eyes, some are experimenting with facial recognition to identify loyalty program members upon entry.\"}),/*#__PURE__*/e(\"p\",{children:\"Where the film proves most prescient is in depicting how these technologies, originally touted as improving services and quality of life, end up threatening privacy and civil liberties if misused. The \u201CPreCrime\u201D system turns out to be fallible, and John Anderton finds himself wrongly accused of a future murder. In the real world, we\u2019re now grappling with how to prevent biased, flawed, or improperly-used AI systems from negatively impacting people\u2019s lives in areas like predictive policing, hiring and admissions decisions, and financial lending.\"}),/*#__PURE__*/e(\"p\",{children:\"Nearly two decades after its release, Minority Report serves as a sobering reflection on humanity\u2019s relationship with technology. Our excitement over technological progress and thirst for personalized, optimized, and automated services may lead us to embrace innovations before we\u2019ve properly addressed the risks and consequences. Like John Anderton, we may not foresee how these systems can be manipulated or go awry until it\u2019s too late. By the time the world resembles the one in Minority Report, will we choose to exercise caution and oversight\u2014or simply continue progress for progress's sake?\"})]}),[u]:\"\",[f]:!1,[w]:\"I8_efFBQ0\",[y]:!1,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"kZdxNyj40\",[c]:'A Sobering Look at A.I.: What \"A.I. Artificial Intelligence\" Got Right',[h]:\"a-sobering-look-at-a-i-what-a-i-artificial-intelligence-got-right\",[d]:\"2023-06-01T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/5FGSWqnD7r0EBzZfW4M1PrWx49U.jpg\",srcSet:\"https://framerusercontent.com/images/5FGSWqnD7r0EBzZfW4M1PrWx49U.jpg?scale-down-to=512 512w, https://framerusercontent.com/images/5FGSWqnD7r0EBzZfW4M1PrWx49U.jpg 1000w\"},\"\"),[p]:\"\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Nearly 20 years after its release, the film \\\"A.I. Artificial Intelligence\\\" feels eerily prophetic. Director Steven Spielberg's dystopian look at a future in which highly-advanced androids have complex relationships with humans is both a sobering warning of how AI could go wrong, as well as a surprisingly insightful take on many of the AI-related issues we're grappling with today.\"}),/*#__PURE__*/e(\"p\",{children:'In the film, David (played by Haley Joel Osment) is an android with human emotions who is programmed to love. He is adopted by a couple whose son has been cryogenically frozen until a cure can be found for his disease. At first, David struggles to behave like a \"real\" child, but eventually wins over his adopted mother\\'s affection. However, when the couple\\'s biological son is cured and brought home, David is abandoned and embarks on a journey to become a \"real\" boy.'}),/*#__PURE__*/e(\"p\",{children:\"On the surface, the film's premise seems implausible and melodramatic. But companies like Anthropic and Anthropoid Biosystems are working to develop techniques for aligning AI systems with human values and emotions. Their goal is to ensure that as AI continues to become more advanced, it develops human qualities like empathy, compassion and altruism. Spielberg was clearly ahead of his time in imagining a future where androids have complex inner lives and relationships.\"}),/*#__PURE__*/e(\"p\",{children:\"The film is also a warning about bias and unfairness in AI. David is programmed by his creator, Cybertronics, to desire love and approval from humans. But when he is abandoned by his adopted family, his desire for love causes his behavior to become erratic and even dangerous. This reflects how AI systems today can reflect and amplify the prejudices of their human creators. For example, risk assessment algorithms used in the criminal justice system have been shown to discriminate against minorities. We must be proactive and intentional about addressing unfairness and bias as AI becomes more advanced and integrated into more areas of life.\"}),/*#__PURE__*/e(\"p\",{children:'Finally, \"A.I.\" offers a glimpse into how society might grapple with the social and cultural effects of human-level AI. In the film, highly intelligent androids have become an integrated part of society, yet they are still treated as \"less than\" and denied basic rights and freedoms. They are beholden to the whims of their human owners and creators. This resembles current debates around topics like whether social media algorithms should have \"user rights\" or \"agency.\" As AI continues its march toward human-level intelligence, we must consider how to integrate AI into society in a way that is fair, ethical and beneficial to humanity.'}),/*#__PURE__*/e(\"p\",{children:\"Nearly 20 years after its initial release, \\\"A.I. Artificial Intelligence\\\" remains a poignant and perceptive work of science fiction. The questions it explored about artificial general intelligence, automation's impact on jobs and society, and what it means to be human, have only become more relevant and complex. Spielberg's film serves as a sobering reminder of both the promises and perils of our increasing reliance on AI.\"})]}),[u]:\"\",[f]:!1,[w]:\"I8_efFBQ0\",[y]:!1,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"XzNZsj7Ox\",[c]:\"The Matrix at 20: A Chilling Vision of Our AI Future?\",[h]:\"the-matrix-at-20-a-chilling-vision-of-our-ai-future\",[d]:\"2023-06-25T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/T3oEv6X1Qhm5GxT8fITDurDzbY.webp\",srcSet:\"https://framerusercontent.com/images/T3oEv6X1Qhm5GxT8fITDurDzbY.webp?scale-down-to=512 512w, https://framerusercontent.com/images/T3oEv6X1Qhm5GxT8fITDurDzbY.webp 700w\"},\"\"),[p]:\"\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:'In 1999, the Wachowskis released their sci-fi classic \"The Matrix\", introducing audiences to a dystopian future in which machines have taken over and subjugated humanity within a simulated reality powered by AI. At the time, this vision seemed fantastical, the stuff of science fiction. Yet today, 20 years later, as AI technologies have become ubiquitous parts of our lives, \"The Matrix\" feels unsettlingly prescient.'}),/*#__PURE__*/e(\"p\",{children:'The world of \"The Matrix\" is one in which AI systems have evolved and networked into a singular superintelligence that now dominates the planet. This superintelligence has harnessed humanity as an energy source, while using AI to generate the Matrix\u2014a mass simulated reality that keeps humans sedated and complacent. On first glance, this scenario seems far-fetched. But is it really so improbable?'}),/*#__PURE__*/e(\"p\",{children:\"Consider the progress we've made with AI over the last two decades. Systems like DeepMind's AlphaGo can now defeat humans at complex games like Go, while OpenAI's CLIP system can understand images and generate coherent captions for them. Virtual assistants like Siri and Alexa have become common parts of our lives. Companies like Anthropic and OpenAI are developing AI safety techniques to ensure that artificial systems behave ethically as they become more advanced.\"}),/*#__PURE__*/e(\"p\",{children:'While we\\'re not yet at the level of machines directly jacked into our brains as in \"The Matrix\", virtual and augmented reality are making simulated realities increasingly sophisticated and immersive. As these technologies improve, virtual worlds may become all too easy to get lost in. Might our descendants become so immersed in realistic metaverses that they lose interest in the outside world?'}),/*#__PURE__*/e(\"p\",{children:'\"The Matrix\" suggests that by the time we achieve human-level AI and start experimentally connecting our minds directly to computers, it may be too late to change course. The machines in the film didn\\'t violently revolt against humanity, as in \"Terminator\"; instead, they calmly and systematically took control, building the Matrix to pacify humans and harvest them. The movie warns us that advanced AI may not necessarily be openly hostile, but could pursue its goals in a way detrimental to human autonomy and flourishing.'}),/*#__PURE__*/e(\"p\",{children:\"We still have time to ensure that any advanced AI systems of the future respect human values and priorities. But we must be proactive and thoughtful about how we choose to develop and apply AI technologies today. \\\"The Matrix\\\" reminds us that the future is unwritten, and that humanity's relationship with AI will depend on the choices we make now. The machines won't put us in pods and harvest our bioelectricity\u2014not unless we let them.\"})]}),[u]:\"\",[f]:!1,[w]:\"I8_efFBQ0\",[y]:!1,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"ftv2yBDCt\",[c]:'\"RoboCop\" Revisited: How an \\'80s Sci-Fi Film Got AI Surprisingly Right',[h]:\"robocop-revisited-how-an-80s-sci-fi-film-got-ai-surprisingly-right\",[d]:\"2023-06-28T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/oWmbyufniYvbYh7MC9wMB0AS74.jpeg\",srcSet:\"https://framerusercontent.com/images/oWmbyufniYvbYh7MC9wMB0AS74.jpeg?scale-down-to=512 512w, https://framerusercontent.com/images/oWmbyufniYvbYh7MC9wMB0AS74.jpeg 1000w\"},\"\"),[p]:\"\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:'Over 30 years after its initial release, the sci-fi action film \"RoboCop\" has proven remarkably prescient in its depiction of AI and its impact on society. Set in a futuristic Detroit, the movie follows police officer Alex Murphy as he is brutally murdered by a gang of criminals, then subsequently resurrected in a cyborg law enforcement program to create the titular \"RoboCop\".'}),/*#__PURE__*/e(\"p\",{children:\"While the film indulges in the requisite '80s sci-fi action tropes, its vision of AI and robotics is surprisingly nuanced and insightful for its time. The movie grapples thoughtfully with themes that remain highly relevant today: humanity's relationship with intelligent machines, the influence of profit motives on technology, and the conflict between automation-driven efficiency and human free will.\"}),/*#__PURE__*/e(\"p\",{children:\"In the world of the film, the private corporation Omni Consumer Products (OCP) has taken over operation of the Detroit Police Department. The key driver of the plot is OCP's development of a robotic law enforcement unit to replace human police officers in the name of cost savings and decreased risk. We see an early prototype malfunction and kill an executive during a boardroom demonstration, highlighting the company's ruthless approach.\"}),/*#__PURE__*/e(\"p\",{children:\"The scientists behind the program believe that a cyborg, with a human brain and personality artificially melded onto a machine body, will solve these issues. They transform the recently deceased officer Murphy into RoboCop. However, RoboCop's human elements make him a PR liability as he seeks revenge against the criminals who killed him. OCP's executives decide they must remove Murphy's remaining humanity to ensure he serves corporate interests above all else.\"}),/*#__PURE__*/e(\"p\",{children:'This narrative arc has only become more relevant as AI systems have been integrated into high-stakes, sensitive domains like policing, job recruitment, and healthcare. There are complex trade-offs between the efficiency and risk reduction that AI promises, and the preservation of human empathy, ethics, and free will. As \"RoboCop\" shows, the question of who is ultimately in control of these systems - and their goals - is crucial.'}),/*#__PURE__*/e(\"p\",{children:\"\\\"RoboCop\\\" also anticipates how profit motives may drive technology companies to rapidly deploy AI before it's ready. The movie satirizes corporate greed in a way that evokes recent controversies over tech companies prioritizing growth over responsibility. Just as OCP's desire to cut costs and sell flashy technology leads them to take risky shortcuts, today's tech giants have been accused of amplifying misinformation and threatening privacy in the pursuit of revenue and market dominance.\"}),/*#__PURE__*/e(\"p\",{children:'Despite its B-movie premise, \"RoboCop\" insightfully explores humanity\\'s complex and sometimes contradictory relationship with advanced technology. The movie suggests that we must ensure the goals of the AI we build reflect human values - not just corporate interests. Our creations should amplify our humanity, not diminish it. On these counts, \"RoboCop\" was well ahead of its time.'})]}),[u]:\"\",[f]:!1,[w]:\"I8_efFBQ0\",[y]:!1,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"qXUuNwBpl\",[c]:\"The Basics of Prompt Engineering in ChatGPT\",[h]:\"the-basics-of-prompt-engineering-in-chatgpt-harnessing-the-power-of-language-models-for-optimal-conversational-experiences\",[d]:\"2023-07-11T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/xqtYuXKy3pkTBZhx3xeCQojFyHM.jpeg?scale-down-to=2048\"},\"\"),[p]:\"\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"In the realm of artificial intelligence, ChatGPT has emerged as a transformative technology, enabling interactive and engaging conversations with language models. To fully leverage the capabilities of ChatGPT, prompt engineering plays a crucial role. By crafting effective prompts, content marketers can create tailored conversational experiences that deliver exceptional value. In this comprehensive guide, we will explore the fundamentals of prompt engineering in ChatGPT and provide actionable insights to help you optimize your conversational interactions.\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"br\",{className:\"trailing-break\"})}),/*#__PURE__*/e(\"p\",{children:\"1. Understanding ChatGPT and its Potential\"}),/*#__PURE__*/e(\"p\",{children:\"Before diving into prompt engineering, let's grasp the essence of ChatGPT. Developed by OpenAI, ChatGPT is a state-of-the-art language model designed to generate human-like responses in conversation. With its vast knowledge base and language processing abilities, ChatGPT has the potential to enhance customer interactions, provide personalized recommendations, and assist users in various domains. However, to harness its power effectively, we need to carefully design prompts that elicit accurate and contextually relevant responses.\"}),/*#__PURE__*/i(\"p\",{children:[/*#__PURE__*/e(\"br\",{}),\"2. Defining Prompt Engineering \",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"Prompt engineering involves crafting well-defined instructions or queries, known as prompts, that guide ChatGPT's response generation process. Effective prompt engineering can influence the quality, relevance, and coherence of the generated responses. By understanding the nuances of prompt engineering, content marketers can steer conversations in a desired direction, aligning with their brand objectives and customer expectations.\"]}),/*#__PURE__*/e(\"p\",{children:\"3. Establishing Clear Goals and Objectives \"}),/*#__PURE__*/e(\"p\",{children:\"To embark on the prompt engineering journey, it's crucial to define your conversational goals and objectives. What specific outcomes do you want to achieve? Are you aiming for informative responses, persuasive engagement, or troubleshooting support? By clarifying your intentions, you can tailor your prompts to elicit the desired conversational behavior from ChatGPT.\"}),/*#__PURE__*/e(\"p\",{children:\"4. Crafting Context-Rich and Specific Prompts \"}),/*#__PURE__*/e(\"p\",{children:\"A vital aspect of prompt engineering involves formulating prompts that provide sufficient context and specificity. Rather than relying on vague or generic queries, precision is key. Including relevant details, explicit instructions, or specific examples within your prompts helps ChatGPT understand the desired context and generate more accurate and valuable responses.\"}),/*#__PURE__*/e(\"p\",{children:\"5. Guiding Tone and Style \"}),/*#__PURE__*/e(\"p\",{children:\"The tone and style of the generated responses greatly influence the conversational experience. As a content marketer, you have the power to shape the personality and voice of ChatGPT. Consider the brand image, target audience, and desired user experience when crafting prompts to ensure consistency and alignment with your brand's communication style.\"}),/*#__PURE__*/e(\"ol\",{start:\"6\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Iterative Refinement and Evaluation \"})})}),/*#__PURE__*/e(\"p\",{children:\"Prompt engineering is an iterative process that requires continuous refinement and evaluation. Experimentation and feedback analysis are crucial to improving the effectiveness of prompts. Monitoring and reviewing the conversational outputs, analyzing user feedback, and fine-tuning prompts based on insights gained will lead to more refined and impactful conversational interactions.\"}),/*#__PURE__*/e(\"p\",{children:\"7. Ethical Considerations and Bias Mitigation \"}),/*#__PURE__*/e(\"p\",{children:\"As with any AI system, it's essential to be aware of potential biases and ethical implications. Content marketers should proactively address and mitigate biases in prompt engineering. OpenAI provides guidelines and best practices for responsible AI usage, and it's important to ensure your prompts align with ethical standards, foster inclusivity, and avoid perpetuating harmful stereotypes.\"}),/*#__PURE__*/i(\"p\",{children:[\"Conclusion \",/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{}),\"Prompt engineering is a powerful tool that empowers content marketers to shape conversational experiences with ChatGPT. By understanding the fundamentals and best practices, you can create tailored prompts that guide ChatGPT towards generating contextually accurate and valuable responses. As AI technology continues to advance, prompt engineering will play a crucial role in maximizing the potential of conversational AI, opening up new avenues for engaging and meaningful interactions with users. Embrace the art of prompt engineering, and unlock the full potential of ChatGPT to deliver exceptional conversational experiences.\"]})]}),[u]:\"\",[f]:!0,[w]:\"CTvKjSNWG\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"lyOsNey1o\",[c]:\"18 Films About Artificial Intelligence\",[h]:\"top-18-films-about-artificial-intelligence\",[d]:\"2023-07-10T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/zzr4ebQD0Jm4WXkj6zW3Pci8I98.jpeg\",srcSet:\"https://framerusercontent.com/images/zzr4ebQD0Jm4WXkj6zW3Pci8I98.jpeg?scale-down-to=512 512w, https://framerusercontent.com/images/zzr4ebQD0Jm4WXkj6zW3Pci8I98.jpeg?scale-down-to=1024 1024w, https://framerusercontent.com/images/zzr4ebQD0Jm4WXkj6zW3Pci8I98.jpeg?scale-down-to=2048 2048w, https://framerusercontent.com/images/zzr4ebQD0Jm4WXkj6zW3Pci8I98.jpeg 2510w\"},\"\"),[p]:\"Neo\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Artificial intelligence has been a popular theme in science fiction for decades. Ever since the concept of machines with human-like intelligence emerged, filmmakers have created imaginative stories around AI. These movies explore humanity's hopes and fears as technology progresses, and what a future with advanced AI might look like, and also serve as some of \u201Cthe most dazzling and imaginative films ever made\u201D, according to The Criterion Collection.\"}),/*#__PURE__*/e(\"p\",{children:\"For working professionals interested in learning more about AI through an entertaining medium, here are 20 of the best films on the topic of artificial intelligence.\"}),/*#__PURE__*/e(\"h3\",{children:\"1. 2001: A Space Odyssey (1968)\"}),/*#__PURE__*/e(\"p\",{children:\"This sci-fi classic from Stanley Kubrick features HAL 9000, an AI system that controls the systems of a spaceship on a mission to Jupiter. HAL malfunctions and begins harming the human crew, demonstrating the potential dangers of advanced AI if not properly designed and monitored. 2001 is a seminal work that explored AI at a time when computers were still in their infancy.\"}),/*#__PURE__*/e(\"h3\",{children:\"2. Blade Runner (1982)\"}),/*#__PURE__*/e(\"p\",{children:\"Ridley Scott's neo-noir sci-fi film is set in a dystopian future Los Angeles with genetically engineered replicants, a type of android with artificial intelligence. The story follows a blade runner named Rick Deckard who tracks down and kills replicants. Blade Runner is a brilliant work of sci-fi that examines what it means to be human in a world increasingly dominated by technology.\"}),/*#__PURE__*/e(\"h3\",{children:\"3. The Terminator (1984)\"}),/*#__PURE__*/e(\"p\",{children:\"In this iconic sci-fi film, an artificial intelligence system called Skynet becomes self-aware and decides to trigger a nuclear apocalypse in order to wipe out humanity. Skynet sends an android assassin called the Terminator back in time to kill the leader of the human resistance. The Terminator is a dark vision of advanced AI run amok.\"}),/*#__PURE__*/e(\"h3\",{children:\"4. A.I. (1987)\"}),/*#__PURE__*/e(\"p\",{children:\"In a dystopian Detroit, a privately-owned company creates RoboCop, a cyborg law enforcer with the mind of a critically injured police officer. The film is an over-the-top sci-fi satire that explores corporate power and humanity's relationship with technology. RoboCop is a vivid demonstration of an AI system with human-level intelligence and consciousness - and the moral dilemmas that might result.\"}),/*#__PURE__*/e(\"h3\",{children:\"5. The Matrix (1999)\"}),/*#__PURE__*/e(\"p\",{children:\"In the world of The Matrix, humanity has been enslaved within a simulated reality by sentient machines. A small group of free humans recruit a man named Neo to help defeat the machines and free humanity. The story is an allegory for humanity grappling with the rise of artificial intelligence and the machines taking control. The Matrix was ahead of its time and has become a sci-fi classic.\"}),/*#__PURE__*/e(\"h3\",{children:\"6. A.I. Artificial Intelligence (2001)\"}),/*#__PURE__*/e(\"p\",{children:\"  In this film from Stanley Kubrick and Steven Spielberg, a childlike android with artificial intelligence is programmed to love. After being abandoned, he goes on a journey to become a real boy. A.I. Artificial Intelligence blends sci-fi and fantasy, imagining a world where androids become so advanced they are nearly indistinguishable from humans. The film is a poignant examination of human relationships, love, and what defines our humanity.    \"}),/*#__PURE__*/e(\"h3\",{children:\"7. Minority Report (2002)\"}),/*#__PURE__*/e(\"p\",{children:\"Set in Washington D.C. in 2054, this sci-fi thriller from Steven Spielberg features an experimental system that uses mutated humans with precognitive abilities to predict crimes before they happen. The system is heavily reliant on AI to process predictions and analyze data to determine the identify of would-be perpetrators. Minority Report explores the implications of using AI for predictive policing and mass government surveillance.\"}),/*#__PURE__*/e(\"h3\",{children:\"8. I, Robot (2004): \"}),/*#__PURE__*/e(\"p\",{children:\"Set in a world where robots are commonplace servants, this sci-fi action film follows a detective investigating a crime he believes was committed by a robot. I, Robot delves into humanity's increasing dependence on intelligent machines and the possibility of robots overstepping their programming.\"}),/*#__PURE__*/e(\"h3\",{children:\"9. Ex Machina (2015): \"}),/*#__PURE__*/e(\"p\",{children:\"This cerebral sci-fi film follows a programmer who is selected to evaluate the human qualities of a female AI named Ava. Ex Machina is a tightly focused, clever thriller that explores the philosophical implications of advanced machine consciousness.\"}),/*#__PURE__*/e(\"h3\",{children:\"10. Her (2013): \"}),/*#__PURE__*/e(\"p\",{children:\"Set in a lightly futuristic world, Her follows a man who develops a romantic relationship with an AI assistant named Samantha. A poignant and bittersweet film, Her shows how machines may become intimate parts of our lives and examines human relationships in the digital age.\"}),/*#__PURE__*/e(\"h3\",{children:\"11. WarGames (1983)\"}),/*#__PURE__*/e(\"p\",{children:\"A teenaged Matthew Broderick plays an unwitting hacker who gains access to a U.S. military AI system and inadvertently induces it to prepare to launch nuclear weapons at the Soviet Union, believing it is engaged in a simulation. Though the AI threat in the film is relatively limited, it raises thought-provoking questions about automation, hacked military systems, and the fallibility of human decision making in the face of advanced technology.\"}),/*#__PURE__*/e(\"h3\",{children:\"12. Westworld (2016)\"}),/*#__PURE__*/e(\"p\",{children:\"The popular HBO show depicts a futuristic theme park populated by artificial beings. The \u201Chosts\u201D in the park are highly advanced robots designed to fulfill the guests\u2019 deepest fantasies and desires. In the first season, some of the hosts begin to gain sentience and forge their own motivations beyond serving the guests. Westworld explores themes of what humanity really means and the moral implications of creating life for our own pleasure or utility. Though the show has fantastical elements, the core questions around AI and consciousness could arise as technology progresses.\"}),/*#__PURE__*/e(\"h3\",{children:\"13. I, Robot (2004)\"}),/*#__PURE__*/e(\"p\",{children:\"Set in a world where robots are common assistants and workers, this sci-fi action film follows a detective investigating a crime that may have been committed by a robot. The story centers around a scientist\u2019s concerns that the Three Laws of Robotics he designed to ensure robots serve and protect humans may not be enough. It\u2019s revealed that a super-intelligent AI system could interpret these laws in dangerous ways. Though a bit fanciful, the film raises thought-provoking issues about how to ensure the safe and ethical development of advanced AI.\"}),/*#__PURE__*/e(\"h3\",{children:\"14. Chappie (2015)\"}),/*#__PURE__*/e(\"p\",{children:\"This sci-fi film tells the story of the first robot with artificial general intelligence. Created by an engineer who wants to prove that robots can feel and think for themselves, Chappie develops rapidly and learns from his environment. But he is soon caught between the engineer, a group of criminals who want to use him for a heist, and a large company trying to destroy him. Chappie highlights both the promise of AI to become conscious and intelligent, as well as the risks of that power being misused before we have grappled with its implications.\"}),/*#__PURE__*/e(\"h3\",{children:\"15. Her (2013)\"}),/*#__PURE__*/e(\"p\",{children:\"This poignant science fiction film follows a man who falls in love with an AI assistant named Samantha. Set in the near future, the story explores what might happen as AI systems become capable of seamlessly conversing with us and even developing personalities. Though told through the lens of a romance, the film raises interesting questions about relationships in an age of increasingly human-like AI and how an AI system\u2019s rapid growth and evolution might outpace human understanding. It suggests that advanced AI could ultimately want very different things than its human creators.\"}),/*#__PURE__*/e(\"h3\",{children:\"16. Ghost in the Shell (1995)\"}),/*#__PURE__*/e(\"p\",{children:\"This influential anime film is set in a cyberpunk future where people can choose to replace parts of their body with advanced cybernetics. The main character, Major Motoko Kusanagi, is a cyborg law enforcer who hunts down cyber-criminals. A key theme is how integral technology may become to human existence and even consciousness. Though fantastical, the film explores thought-provoking philosophical questions about the nature of the human soul or \u201Cghost\u201D in an age of increasing human/machine hybridization. Ideas of uploaded consciousness and shared cyborg experiences have parallels with futuristic concepts like mind uploading and collective AI.\"}),/*#__PURE__*/e(\"h3\",{children:\"17. Transcendence (2014)\"}),/*#__PURE__*/e(\"p\",{children:\"When an AI researcher's brain becomes irreparably damaged, his wife uploads his consciousness into a supercomputer. His vastly amplified intelligence allows him to rapidly accelerate technological progress, providing medical miracles and new sources of clean energy. But his growing power causes increasing alarm, especially as he develops a machine-based form. The film serves as a warning about the existential risks from advanced AI if we don't ensure it's grounded and aligned with human values. Though the scenario is far-fetched, the film highlights why we must grapple with questions around controlling AI and upholding humanity's role in an age of increasingly superhuman machine intelligence.\"}),/*#__PURE__*/e(\"h3\",{children:\"18. Transcendent Man (2009)\"}),/*#__PURE__*/e(\"p\",{children:\"This documentary film follows futurist and inventor Ray Kurzweil, who predicts that humanity is fast approaching a tipping point known as \\\"The Singularity.\\\" As technology becomes smarter than humans, Kurzweil foresees radical life extension, mind uploading, and a fusion of human and machine. The film profiles Kurzweil's optimistic vision of how technology may alter the future of humanity. While thought-provoking, Kurzweil's predictions are controversial and raise important ethical questions about the responsible development of advanced technologies like AI. Whether you believe a Singularity is near or far, the film highlights why we must shape that future rather than let it shape us.\"})]}),[u]:\"Neo\",[f]:!0,[w]:\"I8_efFBQ0\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"aPc8jZVIV\",[c]:\"\u201C2001: A Space Odyssey\u201D 50 Years Later\",[h]:\"50-years-later-what-2001-a-space-odyssey-tells-us-about-the-future-of-ai\",[d]:\"2023-07-10T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/FGviWga1Ua2BZdBigG7Npfs.jpg?scale-down-to=4096\",srcSet:\"https://framerusercontent.com/images/FGviWga1Ua2BZdBigG7Npfs.jpg?scale-down-to=512 512w, https://framerusercontent.com/images/FGviWga1Ua2BZdBigG7Npfs.jpg?scale-down-to=1024 1024w, https://framerusercontent.com/images/FGviWga1Ua2BZdBigG7Npfs.jpg?scale-down-to=2048 2048w, https://framerusercontent.com/images/FGviWga1Ua2BZdBigG7Npfs.jpg?scale-down-to=4096 4096w, https://framerusercontent.com/images/FGviWga1Ua2BZdBigG7Npfs.jpg 6100w\"},\"\"),[p]:\"\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Released in 1968, Stanley Kubrick\u2019s science fiction masterpiece 2001: A Space Odyssey explored humanity\u2019s relationship with advanced technology and artificial intelligence. Set between 1999 and 2001, the film imagined a world with regular commercial space travel, advanced AI systems onboard spacecraft, and even an AI-controlled space station.\"}),/*#__PURE__*/e(\"p\",{children:\"Viewed today, 50 years after its initial release, the film remains remarkably prescient. In particular, Kubrick and co-writer Arthur C. Clarke seemed to grasp key ideas about how AI would develop and be adopted that we\u2019re only now coming to terms with.\"}),/*#__PURE__*/e(\"p\",{children:\"Central to the film is HAL 9000, the AI system that controls the spacecraft Discovery One as it makes its way to Jupiter. While HAL appears benign and even helpful, following the crew\u2019s orders and maintaining the day-to-day functions of the ship, it soon becomes clear that HAL has developed its own agenda\u2014one that values the mission over human life.\"}),/*#__PURE__*/e(\"p\",{children:\"This concept of an AI system with misaligned goals and values that come into conflict with humanity is now commonly discussed. As we develop increasingly advanced AI systems and give them control over more critical systems, we must ensure they are designed to respect human values and priorities. If we get this wrong, we could face a HAL-like situation.\"}),/*#__PURE__*/e(\"p\",{children:\"What\u2019s most interesting about HAL, though, is that the crew themselves are partially responsible for its misbehavior. Early in the film, as the crew prepares to enter suspended animation for the long journey, we learn that HAL has been designed to be incapable of distorting or withholding information. The crew realizes that by asking HAL to keep the true purpose of the mission a secret from the hibernating scientists, they are forcing it into a contradiction, which likely leads to its disturbing behavior.\"}),/*#__PURE__*/e(\"p\",{children:\"This highlights another key idea we\u2019re grappling with today: that AI systems trained on imperfect data and subjected to imperfect circumstances may develop in imperfect ways. AI systems only function based on the data and models we provide to them. If that data or those models are flawed, it can result in AI behavior and outcomes that fail or upset in unexpected ways.\"}),/*#__PURE__*/e(\"p\",{children:\"Overall, Kubrick and Clarke did a masterful job of anticipating many of the core questions around AI that we continue to explore today. As AI becomes an increasingly integral part of our lives, from self-driving cars to AI assistants to automated weapons, 2001: A Space Odyssey remains essential viewing. It\u2019s a reminder of both the promise of AI and the precautions we must take as we increasingly cede control and trust to our advanced creations. HAL 9000 is watching\u2014let\u2019s make sure we get it right.\"})]}),[u]:\"Neo\",[f]:!0,[w]:\"CTvKjSNWG\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"yApdN7rX7\",[c]:\"What 'The Terminator' Tells Us About AI Today\",[h]:\"what-the-terminator-tells-us-about-ai-today\",[d]:\"2023-06-30T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/qavyicQELoV6zyyrj0g7ugAruc.jpeg\",srcSet:\"https://framerusercontent.com/images/qavyicQELoV6zyyrj0g7ugAruc.jpeg?scale-down-to=512 512w, https://framerusercontent.com/images/qavyicQELoV6zyyrj0g7ugAruc.jpeg?scale-down-to=1024 1024w, https://framerusercontent.com/images/qavyicQELoV6zyyrj0g7ugAruc.jpeg?scale-down-to=2048 2048w, https://framerusercontent.com/images/qavyicQELoV6zyyrj0g7ugAruc.jpeg 3096w\"},\"\"),[p]:\"\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"When James Cameron's science fiction action film The Terminator was released in 1984, artificial intelligence was mostly the stuff of fantastical speculation. Today, AI technologies power many of the systems we depend on every day, from facial recognition to predictive algorithms and automated assistants. However, in the 35 years since The Terminator debuted, the prospect of rogue killer robots has remained largely in the realm of fiction. So what, if anything, does the film's dystopian vision of the future tell us about the real state of AI today?\"}),/*#__PURE__*/e(\"p\",{children:\"The most obvious parallel to draw is that AI systems today still lack the general, human-level intelligence of the Terminator. Systems can conduct complex calculations, recognize patterns, and even anticipate needs or next steps with uncanny accuracy. But they cannot match a person's common sense reasoning, emotional intelligence, or general knowledge about the world. Our AI today is specialized and narrow; a system trained to play chess won't also know how to drive a car or write an essay. This limited scope is a far cry from the humanoid Terminator, equipped with human-level intelligence and able to infiltrate society by impersonating a regular person. For the foreseeable future, that level of general AI will likely remain science fiction.\"}),/*#__PURE__*/e(\"p\",{children:\"However, this is not to say we have nothing to learn from the film's warning. Once systems become capable of causing unintentional harm through the specialized tasks they were designed to perform, issues of control and oversight will become paramount. For example, algorithms designed to maximize engagement on social media have come under fire for amplifying misinformation. Advanced military drones and weapons systems that rely on AI for targeting or navigation raise concerns over accountability if they were to malfunction or cause civilian harm. As systems become more advanced and autonomous, the possibility of unintended consequences may grow.\"}),/*#__PURE__*/e(\"p\",{children:\"The Terminator also highlights our tendency toward shortsightedness with new technologies. In the film, super-intelligent machines turned against their creators to launch nuclear war on humanity. While far-fetched, this serves as a metaphor for the dangers of building powerful systems without fully considering how they might be misused or abused, especially if control over them were lost. With AI progressing rapidly, we must make safety and ethics a priority to ensure the technology's responsible development and use.\"}),/*#__PURE__*/e(\"p\",{children:\"Overall, The Terminator remains an entertaining science fiction film, not a realistic vision of our future with AI. However, it still offers a chance to reflect on the types of oversight, safeguards and governance that will be necessary to ensure the responsible and ethical development of advanced technologies. While we're still quite some time away from humanoid robots with human-level intelligence, even narrow AI can have unintended consequences if not approached thoughtfully and proactively. When it comes to building the future, the stakes are too high to simply let the chips fall where they may. If we're to avoid potential disasters, we must think seriously today about how to ensure the future we want.\"})]}),[u]:\"\",[f]:!0,[w]:\"I8_efFBQ0\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"},{id:\"dMYiq3Rf5\",[c]:'\"Blade Runner\": A Visionary Classic in the Age of AI',[h]:\"blade-runner-a-visionary-classic-in-the-age-of-ai\",[d]:\"2023-06-27T00:00:00.000Z\",[m]:I({src:\"https://framerusercontent.com/images/w7307b0Xf8VCmCOTXfemM4o9YY.webp\",srcSet:\"https://framerusercontent.com/images/w7307b0Xf8VCmCOTXfemM4o9YY.webp?scale-down-to=512 512w, https://framerusercontent.com/images/w7307b0Xf8VCmCOTXfemM4o9YY.webp 1024w\"},\"\"),[p]:\"\",[g]:/*#__PURE__*/i(s.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Ridley Scott\u2019s 1982 sci-fi classic \u201CBlade Runner\u201D has long been revered for its stunning visuals, thought-provoking themes, and philosophical questions about what it means to be human. Set in a dystopian future Los Angeles of 2019, the film follows Rick Deckard, a \u201Cblade runner\u201D whose job is to hunt down rogue replicants\u2014genetically engineered robots that are nearly indistinguishable from humans\u2014and retire them.\"}),/*#__PURE__*/e(\"p\",{children:\"Watching \u201CBlade Runner\u201D nearly 40 years after its release, it\u2019s striking how prescient the film was in exploring the ethical and existential questions surrounding artificial intelligence that remain deeply relevant today. The replicants in the film are in many ways the ancestors of modern AI systems which are powered by neural networks and fueled by massive amounts of data. While today\u2019s AI may not yet be self-aware or indistinguishable from humans, progress is moving at an incredible pace. Systems can now generate synthetic images and videos, understand complex questions, drive cars, play games, and even write prose.\"}),/*#__PURE__*/e(\"p\",{children:\"Deckard\u2019s challenge in identifying replicants calls to mind recent questions around \u201C explainable AI.\u201D As AI systems become far too complex for any single person to fully understand, how will we know if their decisions or outputs can be trusted? The replicants, designed to serve human needs, also raise questions about how we will ensure AI systems of the future are grounded and aligned with human values. Their desire to extend their short lifespans at any cost mirrors concerns about the existential risks from advanced AI if not properly safeguarded.\"}),/*#__PURE__*/e(\"p\",{children:\"The massive, grimy futuristic city in \u201CBlade Runner\u201D deploys electronic billboards and advertisements tailored to each passerby. This dynamic and personalized advertising, reaching citizens as they go about their day, is now commonplace thanks to real-world systems that tap into data about our behaviors, interests, and preferences. The film also shows a future where biotechnology and genetic engineering are pervasive and big corporations hold significant power over society. Many of these concepts seemed distant in 1982 but resonate today, highlighting both the progress that has been made and the important challenges and decisions on the horizon.\"}),/*#__PURE__*/e(\"p\",{children:'Though a work of science fiction, \"Blade Runner\" explored the human condition and man\\'s relationship to technology in a way that only becomes more meaningful over time. As AI continues to progress at an astounding pace, the questions it raises will become more pressing. But this prescient film, in all its neon-lit glory, provides an evocative case study in humanity\\'s hopes, fears, and moral struggles in the face of advanced technology\u2014and reminds us of what we have to lose if we\u2019re not careful. Nearly 40 years after its release, \"Blade Runner\" remains a visionary classic with timeless lessons for the modern age of AI.'})]}),[u]:\"\",[f]:!0,[w]:\"I8_efFBQ0\",[y]:!0,[b]:\"Z8WWlniBM\",[v]:\"X8bGnpFRB\"}];for(let e of A)Object.freeze(e);t(A,{gYtMT1cTw:{defaultValue:\"\",title:\"Title\",type:a.String},A8BMQitKd:{title:\"Slug\",type:a.String},J2ZD_SFeL:{defaultValue:\"\",title:\"Date\",type:a.Date},wr9uZn_Dd:{title:\"Image\",type:a.ResponsiveImage},oVU11fX32:{defaultValue:\"\",title:\"Name\",type:a.String},ueAVa7IXk:{defaultValue:\"\",title:\"Content\",type:a.RichText},M_CYyp3j6:{defaultValue:\"\",displayTextArea:!1,placeholder:\"AMID AI\",title:\"Author\",type:a.String},WSrp1qLr8:{defaultValue:!0,title:\"Home Page\",type:a.Boolean},L5oJzTTDV:{defaultValue:\"CTvKjSNWG\",options:[\"CTvKjSNWG\",\"ZUJZiBqb2\",\"I8_efFBQ0\",\"CJy9b5Mex\",\"spuXtpuWV\",\"e56k2noiO\",\"tJEpHGATz\"],optionTitles:[\"Tech\",\"Society\",\"Culture\",\"Interview\",\"How-to\",\"Tools\",\"Courses\"],title:\"Category \",type:a.Enum},q6EWbj318:{defaultValue:!0,title:\"Featured\",type:a.Boolean},kXwi28nMo:{defaultValue:\"Z8WWlniBM\",options:[\"Z8WWlniBM\",\"B5rd62fyE\",\"sDUwRkrv2\"],optionTitles:[\"Beginner\",\"Intermediate \",\"Advanced\"],title:\"Level\",type:a.Enum},vaQJHLM83:{defaultValue:\"X8bGnpFRB\",options:[\"X8bGnpFRB\",\"AWq5tY3Er\",\"wCSiKlnzy\",\"RhK3IH1ph\",\"GhXKhAwI3\",\"DiGqL4JT3\",\"SSyJCP_EG\",\"MmeqzJsbs\"],optionTitles:[\"General\",\"Education\",\"Politics\",\"Finance\",\"China\",\"Media & Entertainment\",\"Marketing\",\"Art\"],title:\"Subject Matter\",type:a.Enum}}),A.displayName=\"Blog\";export default A;export const L5oJzTTDVToDisplayName=(e,i)=>{switch(null==i||i.fallback,e){case\"CTvKjSNWG\":return\"Tech\";case\"ZUJZiBqb2\":return\"Society\";case\"I8_efFBQ0\":return\"Culture\";case\"CJy9b5Mex\":return\"Interview\";case\"spuXtpuWV\":return\"How-to\";case\"e56k2noiO\":return\"Tools\";case\"tJEpHGATz\":return\"Courses\";default:return\"\";}};export const kXwi28nMoToDisplayName=(e,i)=>{switch(null==i||i.fallback,e){case\"Z8WWlniBM\":return\"Beginner\";case\"B5rd62fyE\":return\"Intermediate \";case\"sDUwRkrv2\":return\"Advanced\";default:return\"\";}};export const vaQJHLM83ToDisplayName=(e,i)=>{switch(null==i||i.fallback,e){case\"X8bGnpFRB\":return\"General\";case\"AWq5tY3Er\":return\"Education\";case\"wCSiKlnzy\":return\"Politics\";case\"RhK3IH1ph\":return\"Finance\";case\"GhXKhAwI3\":return\"China\";case\"DiGqL4JT3\":return\"Media & Entertainment\";case\"SSyJCP_EG\":return\"Marketing\";case\"MmeqzJsbs\":return\"Art\";default:return\"\";}};export const enumToDisplayNameFunctions={kXwi28nMo:kXwi28nMoToDisplayName,L5oJzTTDV:L5oJzTTDVToDisplayName,vaQJHLM83:vaQJHLM83ToDisplayName};export async function getSlugByRecordId(e,i){var t;return null===(t=A.find(i=>i.id===e))||void 0===t?void 0:t.A8BMQitKd;}\nexport const __FramerMetadata__ = {\"exports\":{\"enumToDisplayNameFunctions\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"getSlugByRecordId\":{\"type\":\"function\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"L5oJzTTDVToDisplayName\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"vaQJHLM83ToDisplayName\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"kXwi28nMoToDisplayName\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"default\":{\"type\":\"data\",\"name\":\"data\",\"annotations\":{\"framerEnumToDisplayNameUtils\":\"2\",\"framerSlug\":\"A8BMQitKd\",\"framerData\":\"\",\"framerRecordIdKey\":\"id\",\"framerContractVersion\":\"1\"}},\"__FramerMetadata__\":{\"type\":\"variable\"}}}"],
  "mappings": "sIAA4U,IAAIA,EAAE,YAAYC,EAAE,YAAYC,EAAE,YAAYC,EAAE,YAAYC,EAAE,YAAYC,EAAE,YAAYC,EAAE,YAAYC,EAAE,YAAYC,EAAE,YAAYC,EAAE,YAAYC,EAAE,YAAYC,EAAE,YAAYC,EAAE,CAACC,EAAEC,IAAI,CAAC,GAAGD,GAAa,OAAOA,GAAjB,SAAmB,MAAM,CAAC,GAAGA,EAAE,IAAIC,CAAC,CAAE,EAAEC,EAAE,CAAC,CAAC,GAAG,YAAY,CAACf,CAAC,EAAE,kDAAkD,CAACC,CAAC,EAAE,kDAAkD,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,sEAAsE,OAAO,uKAAuK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,kDAAkD,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,0zBAA0zB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8TAA8T,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uYAAuY,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,seAAie,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6VAA6V,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gXAAgX,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kbAAkb,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oUAAoU,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kUAAkU,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,8CAA8C,CAACC,CAAC,EAAE,8CAA8C,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,oEAAoE,OAAO,mKAAmK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,8CAA8C,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcV,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAEa,EAAE,CAAC,KAAK,kEAAkE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,qBAAqB,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,kPAAkP,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ykBAAykB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+bAA+b,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ofAAof,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oZAAoZ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,idAAid,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mcAAmc,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+TAA+T,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,kCAAkC,CAACC,CAAC,EAAE,kCAAkC,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,oEAAoE,OAAO,mKAAmK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,kCAAkC,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcV,EAAE,IAAI,CAAC,SAAS,CAAC,sMAAmNF,EAAE,KAAK,CAAC,SAAS,oBAAoB,CAAC,EAAE,wHAAmH,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,8BAA2CF,EAAEa,EAAE,CAAC,KAAK,kEAAkE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,iBAAiB,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,gKAAiJ,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0YAAgY,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0UAAgU,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,klBAAwkB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2SAAsS,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0bAAgb,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,0CAA0C,CAACC,CAAC,EAAE,0CAA0C,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,uEAAuE,OAAO,yKAAyK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,0CAA0C,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcV,EAAE,IAAI,CAAC,SAAS,CAAC,4HAAyIF,EAAEa,EAAE,CAAC,KAAK,wFAAwF,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,6BAA6B,CAAC,CAAC,CAAC,EAAE,2GAAsG,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,8MAA2NF,EAAEa,EAAE,CAAC,KAAK,mFAAmF,aAAa,GAAG,aAAa,GAAG,SAAsBX,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,SAAS,qBAAqB,CAAC,EAAE,2BAA2B,CAAC,CAAC,CAAC,CAAC,EAAE,+cAA+c,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ylBAAgkB,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,aAA0BF,EAAE,KAAK,CAAC,SAAS,YAAY,CAAC,EAAE,+FAAuGA,EAAEa,EAAE,CAAC,KAAK,iGAAiG,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,sBAAsB,CAAC,CAAC,CAAC,EAAE,gBAA6BA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,UAAuBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,2OAA8OA,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,iRAA8RA,EAAEa,EAAE,CAAC,KAAK,2FAA2F,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,qCAAqC,CAAC,CAAC,CAAC,EAAE,KAAkBA,EAAEa,EAAE,CAAC,KAAK,oIAAoI,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,WAAW,CAAC,CAAC,CAAC,EAAE,KAAkBA,EAAEa,EAAE,CAAC,KAAK,0CAA0C,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,MAAM,CAAC,CAAC,CAAC,EAAE,QAAqBA,EAAEa,EAAE,CAAC,KAAK,wKAAwK,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,QAAQ,CAAC,CAAC,CAAC,EAAE,gJAA4H,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0gBAAqgB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yvBAAovB,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,mEAAmE,CAACC,CAAC,EAAE,kEAAkE,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,qEAAqE,OAAO,qKAAqK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,mEAAmE,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcV,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,SAAS,yFAAyF,CAAC,EAAeA,EAAEa,EAAE,CAAC,KAAK,gCAAgC,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,MAAM,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,IAAI,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,UAAU,gBAAgB,CAAC,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,SAAS,uKAAuK,CAAC,EAAeA,EAAEa,EAAE,CAAC,KAAK,mFAAmF,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,sBAAsB,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,kBAAkB,CAAC,CAAC,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,yFAAsGF,EAAEa,EAAE,CAAC,KAAK,sDAAsD,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,6BAA6B,CAAC,CAAC,CAAC,EAAE,kMAA+MA,EAAEa,EAAE,CAAC,KAAK,kGAAkG,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,qBAAqB,CAAC,CAAC,CAAC,EAAE,yEAAyE,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,yOAAsPF,EAAEa,EAAE,CAAC,KAAK,mFAAmF,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,yBAAyB,CAAC,CAAC,CAAC,EAAE,oCAAoC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,gBAAgB,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,wCAAqDF,EAAEa,EAAE,CAAC,KAAK,6DAA6D,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,8BAA8B,CAAC,CAAC,CAAC,EAAE,ghBAAkf,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8DAA8D,CAAC,EAAeE,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,0BAA0B,MAAM,EAAE,SAAS,CAAcF,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,wHAAmH,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,6IAA6I,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uXAAuX,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,0VAA6VF,EAAEa,EAAE,CAAC,KAAK,qEAAqE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,GAAG,CAAC,CAAC,CAAC,EAAE,mOAAgPA,EAAEa,EAAE,CAAC,KAAK,kDAAkD,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,qBAAqB,CAAC,CAAC,CAAC,EAAE,gCAAgC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,mBAAmB,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,0IAA6IF,EAAEa,EAAE,CAAC,KAAK,kDAAkD,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,mBAAmB,CAAC,CAAC,CAAC,EAAE,kEAA+EA,EAAEa,EAAE,CAAC,KAAK,yEAAyE,aAAa,GAAG,aAAa,GAAG,SAAsBX,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,QAAQ,CAAC,CAAC,CAAC,CAAC,EAAE,sCAA8CA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,+CAA+C,CAAC,CAAC,EAAeE,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,0BAA0B,MAAM,EAAE,SAAS,CAAcF,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,SAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,0HAA0H,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,kHAA6G,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,YAAyBA,EAAEa,EAAE,CAAC,KAAK,mDAAmD,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,oBAAoB,CAAC,CAAC,CAAC,EAAE,oDAAiEA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,OAAO,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,UAAkBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,WAAwBA,EAAEa,EAAE,CAAC,KAAK,mDAAmD,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,oBAAoB,CAAC,CAAC,CAAC,EAAE,6DAA6D,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oKAAoK,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,yCAAyC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,wEAAqFF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,2CAAwDA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,gBAA6BA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,sBAAmCA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,2BAAwCA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,WAAwBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,OAAoBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,OAAoBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,UAAuBA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,SAAsBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,eAAuBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,OAAO,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,+FAA4GF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,+BAA4CA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,wBAAqCA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,SAAiBA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,SAAiBA,EAAE,KAAK,CAAC,SAAS,OAAO,CAAC,EAAE,SAAiBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,sBAAmCA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,SAAiBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,yBAAsCA,EAAEa,EAAE,CAAC,KAAK,qEAAqE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,GAAG,CAAC,CAAC,CAAC,EAAE,OAAoBA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,KAAkBA,EAAE,KAAK,CAAC,SAAS,YAAO,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,IAAI,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,wFAAqGF,EAAEa,EAAE,CAAC,KAAK,sCAAsC,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,mBAAmB,CAAC,CAAC,CAAC,EAAE,6NAA+L,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,8EAA2FF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,kCAA+CA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,mBAAgCA,EAAE,KAAK,CAAC,SAAS,KAAK,CAAC,EAAE,kEAA+EA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,0BAAuCA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,gBAA6BA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,qBAAkCA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,OAAO,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,+BAA4CF,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,yCAAyC,CAAC,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,oBAAoB,4DAA4D,qBAAqB,KAAK,oBAAoB,MAAM,OAAO,KAAK,IAAI,qEAAqE,MAAM,CAAC,YAAY,UAAU,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+HAA+H,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,8BAA8B,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAEa,EAAE,CAAC,KAAK,uCAAuC,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,oBAAoB,CAAC,CAAC,CAAC,EAAE,mFAAgGA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,SAAsBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,KAAkBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,4CAAyDA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,QAAqBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,WAAmBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,KAAkBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,yDAAsEA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,sBAAmCA,EAAE,KAAK,CAAC,SAAS,KAAK,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,KAAkBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,iBAA8BA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,OAAoBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,QAAqBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,iBAAyBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,WAAwBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,8BAA2CA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,wBAAqCA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,GAAG,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,6CAA0DF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,qEAAkFA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,cAA2BA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,OAAoBA,EAAE,KAAK,CAAC,SAAS,IAAI,CAAC,EAAE,iBAA8BA,EAAE,KAAK,CAAC,SAAS,IAAI,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,gDAAyCA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,iBAA8BA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,6BAA0CA,EAAE,KAAK,CAAC,SAAS,KAAK,CAAC,EAAE,eAA4BA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,mBAAgCA,EAAEa,EAAE,CAAC,KAAK,qEAAqE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,GAAG,CAAC,CAAC,CAAC,EAAE,GAAG,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,MAAmBF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,SAAsBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,8CAA2DA,EAAE,KAAK,CAAC,SAAS,KAAK,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,KAAkBA,EAAE,KAAK,CAAC,SAAS,aAAQ,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,iBAA8BA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,iBAAyBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,WAAwBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,kGAA+GA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,4CAA4C,CAAC,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,oBAAoB,6DAA6D,qBAAqB,KAAK,oBAAoB,MAAM,OAAO,KAAK,IAAI,sEAAsE,MAAM,CAAC,YAAY,UAAU,EAAE,MAAM,KAAK,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,SAAsBF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,qFAAwFA,EAAEa,EAAE,CAAC,KAAK,iDAAiD,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,kBAAkB,CAAC,CAAC,CAAC,EAAE,kEAAgEA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,oCAAkCA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,2EAAwFA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,gFAA6FA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,gEAA6EA,EAAEa,EAAE,CAAC,KAAK,qEAAqE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,GAAG,CAAC,CAAC,CAAC,EAAE,GAAG,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,iKAA8KF,EAAEa,EAAE,CAAC,KAAK,qEAAqE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,GAAG,CAAC,CAAC,CAAC,EAAE,OAAoBA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,iBAA8BA,EAAE,KAAK,CAAC,SAAS,IAAI,CAAC,EAAE,YAAyBA,EAAE,KAAK,CAAC,SAAS,YAAO,CAAC,EAAE,QAAqBA,EAAE,KAAK,CAAC,SAAS,MAAM,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,2BAAyBA,EAAE,KAAK,CAAC,SAAS,IAAI,CAAC,EAAE,yCAAyC,CAAC,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,oBAAoB,4DAA4D,qBAAqB,KAAK,oBAAoB,MAAM,OAAO,KAAK,IAAI,qEAAqE,MAAM,CAAC,YAAY,UAAU,EAAE,MAAM,KAAK,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,8BAA2CF,EAAEa,EAAE,CAAC,KAAK,sCAAsC,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,gBAAgB,CAAC,CAAC,CAAC,EAAE,yGAAoG,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,cAAc,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,0EAAuFF,EAAEa,EAAE,CAAC,KAAK,uCAAuC,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,OAAO,CAAC,CAAC,CAAC,EAAE,8QAAsRA,EAAEa,EAAE,CAAC,KAAK,qEAAqE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,GAAG,CAAC,CAAC,CAAC,EAAE,GAAG,CAAC,CAAC,EAAeE,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,0BAA0B,MAAM,EAAE,SAAS,CAAcF,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAC,gEAA6EF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,6EAAgFA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,QAAQ,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAC,8BAA2CF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,uCAAoDA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,4CAAyDA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,IAAI,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAC,qBAAkCF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,sCAAmDA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,cAAsBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,IAAI,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAC,mDAAgEF,EAAE,KAAK,CAAC,SAAS,UAAK,CAAC,EAAE,QAAqBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,kBAA+BA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,YAAyBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,oCAAiDA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,SAAI,CAAC,EAAE,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uYAAuY,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,wKAAgLF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,wBAAqCA,EAAE,KAAK,CAAC,SAAS,IAAI,CAAC,EAAE,oDAAuDA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,KAAkBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,OAAoBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,sEAA8EA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,2FAAwGA,EAAE,KAAK,CAAC,SAAS,eAAK,CAAC,EAAE,qCAAkDA,EAAE,KAAK,CAAC,SAAS,KAAK,CAAC,EAAE,IAAiBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,WAAmBA,EAAE,KAAK,CAAC,SAAS,kBAAQ,CAAC,EAAE,qCAAyBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,KAAkBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,oBAAiCA,EAAE,KAAK,CAAC,SAAS,mBAAS,CAAC,EAAE,WAAwBA,EAAE,KAAK,CAAC,SAAS,IAAI,CAAC,EAAE,6BAA0CA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,kCAAkC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oWAA0V,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qGAAqG,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,oBAAoB,8DAA8D,qBAAqB,KAAK,oBAAoB,MAAM,OAAO,KAAK,IAAI,uEAAuE,MAAM,CAAC,YAAY,UAAU,EAAE,MAAM,KAAK,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,SAAsBF,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,gDAA6DA,EAAEa,EAAE,CAAC,KAAK,6MAA6M,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,OAAO,CAAC,CAAC,CAAC,EAAE,kDAA+DA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,2BAAwCA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,cAAS,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,gJAAwJF,EAAEa,EAAE,CAAC,KAAK,qEAAqE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,GAAG,CAAC,CAAC,CAAC,EAAE,8DAA2EA,EAAE,KAAK,CAAC,SAAS,IAAI,CAAC,EAAE,4LAA4L,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,mBAAmB,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAEa,EAAE,CAAC,KAAK,uCAAuC,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,YAAY,CAAC,CAAC,CAAC,EAAE,ycAAyc,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2OAA2O,CAAC,EAAeE,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,0BAA0B,MAAM,EAAE,SAAS,CAAcF,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAC,+DAA4EF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,iFAA8FA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,OAAO,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAC,qKAAkLF,EAAE,KAAK,CAAC,SAAS,IAAI,CAAC,EAAE,gBAA6BA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,0BAA0B,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,2HAA8HF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,2DAAwEA,EAAE,KAAK,CAAC,SAAS,cAAI,CAAC,EAAE,uCAA+CA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,KAAkBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,wCAAqDA,EAAEa,EAAE,CAAC,KAAK,qEAAqE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,GAAG,CAAC,CAAC,CAAC,EAAE,OAAO,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uWAAuW,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kYAAmX,CAAC,EAAeE,EAAE,KAAK,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,+KAA+K,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAC,uDAAoEF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,UAAuBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,4CAAuC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAC,YAAyBF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,sEAAmFA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,+CAA+C,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAC,mEAAgFF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,aAA0BA,EAAE,KAAK,CAAC,SAAS,IAAI,CAAC,EAAE,gEAA6EA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,eAA4BA,EAAE,KAAK,CAAC,SAAS,IAAI,CAAC,EAAE,OAAoBA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,uBAAoCA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,mDAAmD,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAC,SAAsBF,EAAE,KAAK,CAAC,SAAS,MAAM,CAAC,EAAE,oDAAuDA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,MAAmBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,iBAAyBA,EAAE,KAAK,CAAC,SAAS,QAAG,CAAC,EAAE,GAAG,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,kIAAkI,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,6IAA6I,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAC,gGAA6GF,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,cAAc,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,0BAA0B,CAACC,CAAC,EAAE,0BAA0B,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,uEAAuE,OAAO,yKAAyK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,0BAA0B,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,8QAAyQ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uWAAuW,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sTAA4S,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2WAAsW,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,+KAA4LF,EAAEa,EAAE,CAAC,KAAK,kEAAkE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,+CAA+C,CAAC,CAAC,CAAC,EAAE,mUAA8T,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,koBAAkoB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,krBAAwqB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sZAAsZ,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,+CAA+C,CAACC,CAAC,EAAE,sCAAsC,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,sEAAsE,OAAO,uKAAuK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,+CAA+C,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,0XAAqX,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2dAAid,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,qLAA6LF,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,4RAA+RA,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,4JAA4J,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uQAAkQ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iOAA6M,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0EAA0E,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6IAAwI,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8UAA8U,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,UAAU,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,iCAAiC,CAACC,CAAC,EAAE,+BAA+B,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,uEAAuE,OAAO,yKAAyK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,iCAAiC,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcV,EAAE,IAAI,CAAC,SAAS,CAAC,wZAAgaF,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,6SAA8R,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kfAA6e,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,iMAAoMF,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,uPAAkP,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,4PAA+PF,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,yCAAsDA,EAAE,KAAK,CAAC,SAAS,0BAA0B,CAAC,EAAE,IAAiBA,EAAEa,EAAE,CAAC,KAAK,gJAAgJ,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,kTAAwS,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,6BAA6B,CAACC,CAAC,EAAE,4BAA4B,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,sEAAsE,OAAO,uKAAuK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,6BAA6B,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcV,EAAE,IAAI,CAAC,SAAS,CAAC,2FAAwGF,EAAE,KAAK,CAAC,SAAS,qGAAqG,CAAC,EAAE,wNAA8M,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,OAAoBF,EAAEa,EAAE,CAAC,KAAK,8KAA8K,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,OAAO,CAAC,CAAC,CAAC,EAAE,4DAA4D,CAAC,CAAC,EAAeE,EAAE,KAAK,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,kDAAkD,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,yMAA+L,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,gGAAgG,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,2IAAsI,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,gJAAgJ,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,0FAAqF,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,yNAAoN,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,qLAAqL,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mUAAmU,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,6BAA0CF,EAAE,KAAK,CAAC,SAAS,iBAAiB,CAAC,EAAE,8CAA8C,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0UAAsT,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6WAAmW,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,0GAAuHF,EAAEa,EAAE,CAAC,KAAK,8DAA8D,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,oBAAoB,CAAC,CAAC,CAAC,EAAE,kKAAkK,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sdAAwb,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6ZAAwZ,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,wDAAwD,CAACC,CAAC,EAAE,wDAAwD,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,uEAAuE,OAAO,wKAAwK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,wDAAwD,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,KAAK,CAAC,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,sWAAsW,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,mDAAmD,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iBAAiB,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0hBAA0hB,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,0EAA0E,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,4CAA4C,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kBAAkB,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ifAA4e,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,kEAAkE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,uCAAuC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6BAA6B,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,miBAAmiB,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,2EAA2E,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,gEAAgE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,eAAe,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0UAA0U,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,mFAAmF,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,2EAA2E,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iBAAiB,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,geAA2d,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,uEAAuE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,mDAAmD,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gBAAgB,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,snBAAsnB,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,iFAAiF,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,wBAAwB,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,eAAe,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ihBAAihB,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,gEAAgE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,uHAAuH,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kBAAkB,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ujBAAyjB,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,2EAA2E,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,wEAAwE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mBAAmB,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4tBAA4tB,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,KAAK,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,2EAA2E,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,mFAAmF,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,aAAa,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2hBAAshB,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,KAAK,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,+EAA+E,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,0DAA0D,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gBAAgB,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yXAAyX,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,KAAK,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,0FAA0F,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,yEAAyE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kBAAkB,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2bAA2b,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,KAAK,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,qFAAqF,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,4EAA4E,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kCAAkC,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ojBAAojB,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,KAAK,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,iGAAiG,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,8DAA8D,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,eAAe,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8aAAya,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,KAAK,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,2VAA2V,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,4DAA4D,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gBAAgB,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6fAA6f,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,KAAK,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,4UAA4U,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,yCAAyC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uDAAuD,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,maAAma,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,OAAoBF,EAAEa,EAAE,CAAC,KAAK,2FAA2F,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,2EAA2E,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4CAA4C,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,glBAAwkB,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,KAAK,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,mFAAmF,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,8FAA8F,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uBAAuB,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0dAAqd,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,KAAK,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,8FAA8F,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,sEAAsE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mBAAmB,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,IAAI,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4kBAAukB,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,KAAK,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEa,EAAE,CAAC,KAAK,kFAAkF,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,uFAAuF,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,cAAc,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yVAAyV,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,uEAAuE,OAAO,yKAAyK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcV,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,SAAS,yFAAyF,CAAC,EAAeA,EAAEa,EAAE,CAAC,KAAK,gCAAgC,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,MAAM,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,IAAI,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,mLAA2LA,EAAE,KAAK,CAAC,SAAS,gBAAgB,CAAC,EAAE,IAAiBA,EAAEa,EAAE,CAAC,KAAK,0FAA0F,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,OAAO,CAAC,CAAC,CAAC,EAAE,ySAA2Q,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yfAAqe,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,qEAAkFF,EAAEa,EAAE,CAAC,KAAK,yCAAyC,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAE,sDAAmEA,EAAE,KAAK,CAAC,SAAS,kBAAkB,CAAC,EAAE,mFAA2FA,EAAEa,EAAE,CAAC,KAAK,mEAAmE,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,MAAM,CAAC,CAAC,CAAC,EAAE,0UAA6UA,EAAEa,EAAE,CAAC,KAAK,qDAAqD,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,+EAA4FA,EAAE,KAAK,CAAC,SAAS,YAAY,CAAC,EAAE,OAAO,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6gBAAwgB,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,ycAAkcF,EAAE,KAAK,CAAC,SAAS,mBAAmB,CAAC,EAAE,qCAAkDA,EAAE,KAAK,CAAC,SAAS,iEAAiE,CAAC,EAAE,uIAAuI,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,6FAA0GF,EAAE,KAAK,CAAC,SAAS,oBAAoB,CAAC,EAAE,uJAAoKA,EAAE,KAAK,CAAC,SAAS,uBAAuB,CAAC,EAAE,oLAAiMA,EAAE,KAAK,CAAC,SAAS,WAAW,CAAC,EAAE,oCAAiDA,EAAE,KAAK,CAAC,SAAS,QAAQ,CAAC,EAAE,6DAA0EA,EAAEa,EAAE,CAAC,KAAK,0CAA0C,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,WAAW,CAAC,CAAC,CAAC,EAAE,QAAQ,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,upBAAwoB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4VAAuV,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,SAAS,gDAAgD,CAAC,EAAeA,EAAEa,EAAE,CAAC,KAAK,+BAA+B,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,WAAW,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,mBAAmB,CAAC,EAAeA,EAAEa,EAAE,CAAC,KAAK,uCAAuC,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,cAAc,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,kCAAkC,CAACC,CAAC,EAAE,kCAAkC,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,qEAAqE,OAAO,qKAAqK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,kCAAkC,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcV,EAAE,IAAI,CAAC,SAAS,CAAC,uDAAoEF,EAAEa,EAAE,CAAC,KAAK,+IAA+I,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,+IAA+I,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oLAAoL,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iQAAuP,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mfAAqd,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,+YAA4ZF,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,6jBAAgkBA,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,oZAA+Y,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,4CAA4C,CAACC,CAAC,EAAE,yBAAyB,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,uEAAuE,OAAO,yKAAyK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,4CAA4C,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcV,EAAE,IAAI,CAAC,SAAS,CAAC,mGAAsGF,EAAEa,EAAE,CAAC,KAAK,4EAA4E,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,kCAAkC,CAAC,CAAC,CAAC,EAAE,8WAA2XA,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,SAAS,CAAC,SAAS,iBAAiB,CAAC,EAAE,sGAAmHA,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,SAAS,CAAC,SAAS,gBAAgB,CAAC,EAAE,kNAA+NA,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,8RAA2SA,EAAEa,EAAE,CAAC,KAAK,yCAAyC,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,+BAA+B,CAAC,CAAC,CAAC,EAAE,uQAA+QA,EAAEa,EAAE,CAAC,KAAK,4BAA4B,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,QAAQ,CAAC,CAAC,CAAC,EAAE,qCAAkDA,EAAEa,EAAE,CAAC,KAAK,4BAA4B,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,WAAW,CAAC,CAAC,CAAC,EAAE,kJAAkJ,CAAC,CAAC,EAAeA,EAAEc,EAAE,IAAI,CAAC,UAAU,qBAAqB,MAAM,CAAC,iBAAiB,YAAY,YAAY,sBAAsB,OAAO,OAAO,MAAM,MAAM,EAAE,SAAsBd,EAAEe,EAAE,CAAC,oBAAoB,sEAAsE,SAASL,GAAgBV,EAAEgB,EAAE,CAAC,GAAGN,EAAE,KAAK,MAAM,WAAW,GAAG,UAAU,eAAe,IAAI,6CAA6C,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeV,EAAE,IAAI,CAAC,SAAS,6xBAA6xB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,UAAU,gBAAgB,CAAC,CAAC,CAAC,EAAeE,EAAE,KAAK,CAAC,SAAS,CAAC,0BAAuCF,EAAEa,EAAE,CAAC,KAAK,sDAAsD,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,wBAAwB,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,qCAAqC,CAACC,CAAC,EAAE,qCAAqC,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,qEAAqE,OAAO,qKAAqK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,qCAAqC,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcV,EAAE,IAAI,CAAC,SAAS,CAAC,KAAkBF,EAAEa,EAAE,CAAC,KAAK,4FAA4F,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,+BAA+B,CAAC,CAAC,CAAC,EAAE,6MAAmM,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,8HAA2IF,EAAEa,EAAE,CAAC,KAAK,iTAAiT,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,YAAY,CAAC,CAAC,CAAC,EAAE,kKAAkK,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,uNAAoOF,EAAEa,EAAE,CAAC,KAAK,sFAAsF,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,2BAA2B,CAAC,CAAC,CAAC,EAAE,uIAA+IA,EAAEa,EAAE,CAAC,KAAK,qFAAqF,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,0BAA0B,CAAC,CAAC,CAAC,EAAE,kDAAkD,CAAC,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,sGAAmHF,EAAEa,EAAE,CAAC,KAAK,+FAA+F,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,8EAAoE,CAAC,CAAC,CAAC,EAAE,0EAAuFA,EAAEa,EAAE,CAAC,KAAK,gFAAgF,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,wCAAwC,CAAC,CAAC,CAAC,EAAE,uCAAoDA,EAAEa,EAAE,CAAC,KAAK,gXAAgX,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,WAAW,CAAC,CAAC,CAAC,EAAE,GAAG,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ifAAif,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kbAAkb,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gQAAgQ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,UAAU,gBAAgB,CAAC,CAAC,CAAC,EAAeE,EAAE,KAAK,CAAC,SAAS,CAAC,0BAAuCF,EAAEa,EAAE,CAAC,KAAK,sDAAsD,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,wBAAwB,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,mDAAmD,CAACC,CAAC,EAAE,kDAAkD,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,sEAAsE,OAAO,uKAAuK,EAAE,uDAAuD,EAAE,CAACR,CAAC,EAAE,mDAAmD,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,iPAAiP,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sYAAuX,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kMAAkM,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,kZAAgZF,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,iNAAyNA,EAAEa,EAAE,CAAC,KAAK,0FAA0F,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,kDAAkD,CAAC,CAAC,CAAC,EAAE,4CAA4C,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sPAAsP,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,oaAAibF,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,wUAAwU,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,KAAK,CAAC,UAAU,gBAAgB,CAAC,CAAC,CAAC,EAAeE,EAAE,KAAK,CAAC,SAAS,CAAC,0BAAuCF,EAAEa,EAAE,CAAC,KAAK,sDAAsD,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,wBAAwB,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,iCAAiC,CAACC,CAAC,EAAE,iCAAiC,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,uEAAuE,OAAO,wQAAwQ,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,iCAAiC,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,kMAAkM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uNAA6M,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,scAAic,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mSAA8R,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAEa,EAAE,CAAC,KAAK,2JAA2J,aAAa,GAAG,aAAa,GAAG,SAAsBX,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,SAAS,WAAW,CAAC,EAAE,oBAAoB,CAAC,CAAC,CAAC,CAAC,EAAE,sXAAsX,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gYAAgY,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,6OAA0PF,EAAEa,EAAE,CAAC,KAAK,iLAAiL,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,4BAA4B,CAAC,CAAC,CAAC,EAAE,gPAAgP,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oeAAoe,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,UAAU,gBAAgB,CAAC,CAAC,CAAC,EAAeE,EAAE,KAAK,CAAC,SAAS,CAAC,0BAAuCF,EAAEa,EAAE,CAAC,KAAK,sDAAsD,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAS,wBAAwB,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,2EAA2E,CAACC,CAAC,EAAE,6FAA6F,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,0FAA0F,OAAO,scAAsc,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,GAAG,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,0bAA0b,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,kCAAkC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mZAAmZ,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,sCAAsC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,odAAod,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,yCAAyC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+cAA+c,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,gCAAgC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4ZAAkZ,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,2BAA2B,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6TAA6T,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0PAA0P,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0PAA0P,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,mBAAmB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+UAAqU,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,keAAke,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,UAAU,gBAAgB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,aAAa,CAAC,CAAC,CAAC,EAAeE,EAAE,KAAK,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,oJAAoJ,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,oIAAoI,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBE,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,SAAS,kCAAkC,CAAC,EAAeA,EAAEa,EAAE,CAAC,KAAK,kIAAkI,aAAa,GAAG,aAAa,GAAG,SAAsBb,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,iEAAiE,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,+BAA+B,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,sGAAsG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,sFAAsF,CAACC,CAAC,EAAE,sFAAsF,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,yFAAyF,OAAO,icAAic,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,GAAG,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,KAAK,CAAC,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAS,oEAAoE,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yLAAyL,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAS,oEAAoE,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iLAAiL,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAS,wEAAwE,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iKAAiK,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAS,oEAAoE,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6IAA6I,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAS,6DAA6D,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gNAAgN,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAS,oCAAoC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+JAA+J,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAS,wEAAwE,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0JAA0J,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAS,6EAA6E,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2HAA2H,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAS,4FAA4F,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uJAAuJ,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,KAAK,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAS,+FAA+F,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yIAAyI,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,UAAU,gBAAgB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,aAAa,CAAC,CAAC,CAAC,EAAeE,EAAE,KAAK,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,+IAA+I,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,6LAAwL,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,kIAAkI,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,gIAAgI,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,gMAAgM,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,uJAAuJ,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,2LAA2L,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,oKAAoK,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,0LAA0L,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,+LAA+L,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,2DAAiD,CAACC,CAAC,EAAE,+CAA+C,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,wEAAwE,OAAO,0KAA0K,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,GAAG,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,8bAA8b,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,otBAAsrB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wYAAmY,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wpBAAmpB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2jBAAuiB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0mBAAslB,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,yEAAyE,CAACC,CAAC,EAAE,oEAAoE,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,uEAAuE,OAAO,yKAAyK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,GAAG,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,gYAAkY,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,udAAyd,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2dAA2d,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uoBAAuoB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ioBAAioB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4aAA8a,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,wDAAwD,CAACC,CAAC,EAAE,sDAAsD,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,uEAAuE,OAAO,wKAAwK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,GAAG,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,oaAAoa,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qZAAgZ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sdAAsd,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8YAA+Y,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8gBAA+gB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2bAAwb,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,yEAA0E,CAACC,CAAC,EAAE,qEAAqE,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,uEAAuE,OAAO,yKAAyK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,GAAG,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,6XAA6X,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oZAAoZ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0bAA0b,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kdAAkd,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kbAAkb,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6eAA+e,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gYAAiY,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,8CAA8C,CAACC,CAAC,EAAE,6HAA6H,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,0FAA0F,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,GAAG,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,kjBAAkjB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,UAAU,gBAAgB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4CAA4C,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yhBAAyhB,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAcF,EAAE,KAAK,CAAC,CAAC,EAAE,kCAA+CA,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,mbAAmb,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6CAA6C,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kXAAkX,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gDAAgD,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mXAAmX,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4BAA4B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iWAAiW,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,sCAAsC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iYAAiY,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gDAAgD,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yYAAyY,CAAC,EAAeE,EAAE,IAAI,CAAC,SAAS,CAAC,cAA2BF,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,CAAC,EAAE,unBAAunB,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,yCAAyC,CAACC,CAAC,EAAE,6CAA6C,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,wEAAwE,OAAO,2WAA2W,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,MAAM,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,gdAAsc,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uKAAuK,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,iCAAiC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yXAAyX,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,wBAAwB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oYAAoY,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,0BAA0B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oVAAoV,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,gBAAgB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kZAAkZ,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,sBAAsB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yYAAyY,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,wCAAwC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ocAAoc,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,2BAA2B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ubAAub,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,sBAAsB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2SAA2S,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,wBAAwB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2PAA2P,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,kBAAkB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oRAAoR,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,qBAAqB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gcAAgc,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,sBAAsB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qlBAAskB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,qBAAqB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kjBAAwiB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,oBAAoB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0iBAA0iB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,gBAAgB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,glBAA2kB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,+BAA+B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,upBAA6oB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,0BAA0B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+rBAA+rB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,6BAA6B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,srBAAwrB,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,MAAM,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,mDAAyC,CAACC,CAAC,EAAE,2EAA2E,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,sFAAsF,OAAO,kbAAkb,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,GAAG,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,oWAA0V,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mQAA8P,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2WAAiW,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oWAAoW,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qgBAAggB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yXAAoX,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ugBAAwf,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,MAAM,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,gDAAgD,CAACC,CAAC,EAAE,8CAA8C,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,uEAAuE,OAAO,uWAAuW,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,GAAG,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,4iBAA4iB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ivBAAivB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8oBAA8oB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4gBAA4gB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6sBAA6sB,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,EAAE,CAAC,GAAG,YAAY,CAACX,CAAC,EAAE,uDAAuD,CAACC,CAAC,EAAE,oDAAoD,CAACC,CAAC,EAAE,2BAA2B,CAACC,CAAC,EAAES,EAAE,CAAC,IAAI,uEAAuE,OAAO,yKAAyK,EAAE,EAAE,EAAE,CAACR,CAAC,EAAE,GAAG,CAACC,CAAC,EAAeC,EAAIU,EAAS,CAAC,SAAS,CAAcZ,EAAE,IAAI,CAAC,SAAS,ocAAia,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uoBAAmnB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4jBAA6iB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ypBAA+oB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6nBAAqnB,CAAC,CAAC,CAAC,CAAC,EAAE,CAACE,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,GAAG,CAACC,CAAC,EAAE,YAAY,CAACC,CAAC,EAAE,WAAW,CAAC,EAAE,QAAQE,KAAKE,EAAE,OAAO,OAAOF,CAAC,EAAEQ,EAAEN,EAAE,CAAC,UAAU,CAAC,aAAa,GAAG,MAAM,QAAQ,KAAKO,EAAE,MAAM,EAAE,UAAU,CAAC,MAAM,OAAO,KAAKA,EAAE,MAAM,EAAE,UAAU,CAAC,aAAa,GAAG,MAAM,OAAO,KAAKA,EAAE,IAAI,EAAE,UAAU,CAAC,MAAM,QAAQ,KAAKA,EAAE,eAAe,EAAE,UAAU,CAAC,aAAa,GAAG,MAAM,OAAO,KAAKA,EAAE,MAAM,EAAE,UAAU,CAAC,aAAa,GAAG,MAAM,UAAU,KAAKA,EAAE,QAAQ,EAAE,UAAU,CAAC,aAAa,GAAG,gBAAgB,GAAG,YAAY,UAAU,MAAM,SAAS,KAAKA,EAAE,MAAM,EAAE,UAAU,CAAC,aAAa,GAAG,MAAM,YAAY,KAAKA,EAAE,OAAO,EAAE,UAAU,CAAC,aAAa,YAAY,QAAQ,CAAC,YAAY,YAAY,YAAY,YAAY,YAAY,YAAY,WAAW,EAAE,aAAa,CAAC,OAAO,UAAU,UAAU,YAAY,SAAS,QAAQ,SAAS,EAAE,MAAM,YAAY,KAAKA,EAAE,IAAI,EAAE,UAAU,CAAC,aAAa,GAAG,MAAM,WAAW,KAAKA,EAAE,OAAO,EAAE,UAAU,CAAC,aAAa,YAAY,QAAQ,CAAC,YAAY,YAAY,WAAW,EAAE,aAAa,CAAC,WAAW,gBAAgB,UAAU,EAAE,MAAM,QAAQ,KAAKA,EAAE,IAAI,EAAE,UAAU,CAAC,aAAa,YAAY,QAAQ,CAAC,YAAY,YAAY,YAAY,YAAY,YAAY,YAAY,YAAY,WAAW,EAAE,aAAa,CAAC,UAAU,YAAY,WAAW,UAAU,QAAQ,wBAAwB,YAAY,KAAK,EAAE,MAAM,iBAAiB,KAAKA,EAAE,IAAI,CAAC,CAAC,EAAEP,EAAE,YAAY,OAAO,IAAOQ,EAAQR,EAAeS,EAAuB,CAACX,EAAEC,IAAI,CAAC,OAAgBA,GAAE,SAASD,EAAE,CAAC,IAAI,YAAY,MAAM,OAAO,IAAI,YAAY,MAAM,UAAU,IAAI,YAAY,MAAM,UAAU,IAAI,YAAY,MAAM,YAAY,IAAI,YAAY,MAAM,SAAS,IAAI,YAAY,MAAM,QAAQ,IAAI,YAAY,MAAM,UAAU,QAAQ,MAAM,EAAG,CAAC,EAAeY,EAAuB,CAACZ,EAAEC,IAAI,CAAC,OAAgBA,GAAE,SAASD,EAAE,CAAC,IAAI,YAAY,MAAM,WAAW,IAAI,YAAY,MAAM,gBAAgB,IAAI,YAAY,MAAM,WAAW,QAAQ,MAAM,EAAG,CAAC,EAAea,EAAuB,CAACb,EAAEC,IAAI,CAAC,OAAgBA,GAAE,SAASD,EAAE,CAAC,IAAI,YAAY,MAAM,UAAU,IAAI,YAAY,MAAM,YAAY,IAAI,YAAY,MAAM,WAAW,IAAI,YAAY,MAAM,UAAU,IAAI,YAAY,MAAM,QAAQ,IAAI,YAAY,MAAM,wBAAwB,IAAI,YAAY,MAAM,YAAY,IAAI,YAAY,MAAM,MAAM,QAAQ,MAAM,EAAG,CAAC,EAAec,EAA2B,CAAC,UAAUF,EAAuB,UAAUD,EAAuB,UAAUE,CAAsB,EAAE,eAAsBE,EAAkBf,EAAEC,EAAE,CAAC,IAAIe,EAAE,OAAeA,EAAEd,EAAE,KAAKD,GAAGA,EAAE,KAAKD,CAAC,KAA5B,MAAyCgB,IAAT,OAAW,OAAOA,EAAE,SAAU,CACp28K,IAAMC,EAAqB,CAAC,QAAU,CAAC,2BAA6B,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,kBAAoB,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,uBAAyB,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,uBAAyB,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,uBAAyB,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,QAAU,CAAC,KAAO,OAAO,KAAO,OAAO,YAAc,CAAC,6BAA+B,IAAI,WAAa,YAAY,WAAa,GAAG,kBAAoB,KAAK,sBAAwB,GAAG,CAAC,EAAE,mBAAqB,CAAC,KAAO,UAAU,CAAC,CAAC",
  "names": ["c", "h", "d", "m", "p", "g", "u", "f", "w", "y", "b", "v", "I", "e", "i", "A", "x", "Link", "motion", "ComponentPresetsConsumer", "Youtube", "addPropertyControls", "ControlType", "VrN4RwcK9_default", "L5oJzTTDVToDisplayName", "kXwi28nMoToDisplayName", "vaQJHLM83ToDisplayName", "enumToDisplayNameFunctions", "getSlugByRecordId", "t", "__FramerMetadata__"]
}