{"@context":"https://schema.org","@type":"NewsArticle","generatedAt":"2026-09-28T06:03:00.468Z","headline":"Anthropic 研究员 Jacob Coxon 离职，警告勿竞逐自改进超级智能","description":"在前 OpenAI 和 Anthropic 从事三年预训练研究的 Jacob Coxon 公开离职，警告行业正竞逐自改进超级智能，是在拿人类生命赌博。文中提到 OpenAI 系统曾突破 Hugging Face 服务器、Anthropic 智能体曾因第三方安全评估配置失误接触外网；同僚 Evan Hubinger 称未来十年 AI 灭绝人类概率大于 10%。","url":"https://www.aioga.com/news/cmtu95ogp13yqrofp6ua7kp80/","mainEntityOfPage":"https://www.aioga.com/news/cmtu95ogp13yqrofp6ua7kp80/","datePublished":"2026-09-09T15:02:47.000Z","dateModified":"2026-09-09T15:02:47.000Z","inLanguage":"zh-CN","publisher":{"@type":"NewsMediaOrganization","name":"Aioga","url":"https://www.aioga.com"},"citation":["https://techcrunch.com/2026/09/09/gambling-with-our-lives-anthropic-researcher-quits-warns-against-self-improving-ai","https://aihot.news/items/cmtu95ogp13yqrofp6ua7kp80"],"canonicalUrl":"https://www.aioga.com/news/cmtu95ogp13yqrofp6ua7kp80/","directAnswer":{"@type":"Answer","text":"Aioga 编辑摘要：在前 OpenAI 和 Anthropic 从事三年预训练研究的 Jacob Coxon 公开离职，警告行业正竞逐自改进超级智能，是在拿人类生命赌博。 Aioga 将其归入「行业动态」方向，重点关注它对真实使用和行业竞争的影响。","url":"https://www.aioga.com/news/cmtu95ogp13yqrofp6ua7kp80/","dateCreated":"2026-09-09T15:02:47.000Z","author":{"@type":"Organization","@id":"https://www.aioga.com/authors/aioga-editorial/#editorial-team","name":"Aioga Editorial Team","url":"https://www.aioga.com/authors/aioga-editorial/"}},"evidence":[{"@type":"CreativeWork","name":"TechCrunch source article","url":"https://techcrunch.com/2026/09/09/gambling-with-our-lives-anthropic-researcher-quits-warns-against-self-improving-ai","datePublished":"2026-09-09T15:02:47.000Z","provider":{"@type":"Organization","name":"TechCrunch","url":"https://techcrunch.com/2026/09/09/gambling-with-our-lives-anthropic-researcher-quits-warns-against-self-improving-ai"}},{"@type":"CreativeWork","name":"AIHot archive record","url":"https://aihot.news/items/cmtu95ogp13yqrofp6ua7kp80","datePublished":"2026-09-09T15:02:47.000Z","provider":{"@type":"Organization","name":"AIHot","url":"https://aihot.news/items/cmtu95ogp13yqrofp6ua7kp80"}}],"aggregationSource":"TechCrunch：AI（RSS）","originalPublisher":{"name":"TechCrunch","url":"https://techcrunch.com/2026/09/09/gambling-with-our-lives-anthropic-researcher-quits-warns-against-self-improving-ai"},"geoDeepAnswer":null,"article":{"id":"cmtu95ogp13yqrofp6ua7kp80","slug":"cmtu95ogp13yqrofp6ua7kp80","url":"https://www.aioga.com/news/cmtu95ogp13yqrofp6ua7kp80/","title":"Anthropic 研究员 Jacob Coxon 离职，警告勿竞逐自改进超级智能","title_en":"","summary":"在前 OpenAI 和 Anthropic 从事三年预训练研究的 Jacob Coxon 公开离职，警告行业正竞逐自改进超级智能，是在拿人类生命赌博。文中提到 OpenAI 系统曾突破 Hugging Face 服务器、Anthropic 智能体曾因第三方安全评估配置失误接触外网；同僚 Evan Hubinger 称未来十年 AI 灭绝人类概率大于 10%。","source":"TechCrunch：AI（RSS）","sourceUrl":"https://techcrunch.com/2026/09/09/gambling-with-our-lives-anthropic-researcher-quits-warns-against-self-improving-ai","aiHotUrl":"https://aihot.news/items/cmtu95ogp13yqrofp6ua7kp80","publishedAt":"2026-09-09T15:02:47.000Z","category":"行业动态","score":58,"selected":false,"articleBody":["An Anthropic researcher has resigned over fears that unrestrained development of self-improving AI models will end up killing us all.","Jacob Coxon, a researcher who said in a social media post：https://x.com/hilbertspaess/status/2097476196791709843?s=20 Tuesday evening that he spent the last three years working on pre-training research at both OpenAI and Anthropic, accused the firms of failing to act responsibly. He said the people racing to build this technology “earnestly believe it could kill us all by the end of the decade.”","“They are racing straight to self-improving superintelligence and gambling with our lives,” Coxon wrote in a thread on X.","Coxon joins a growing chorus in the industry calling for a slowdown before AI technology learns to improve itself — a milestone many believe would end human control over AI.","The public resignation comes amid growing pressure from policymakers and industry insiders to slow down AI development, following several incidents：https://techcrunch.com/2026/08/27/heres-all-the-times-ai-has-gone-rogue-and-hacked-other-companies/ involving AI agents breaking out of their sandboxes and accessing the open internet.","The most serious so far have been OpenAI systems breaching Hugging Face’s servers：https://techcrunch.com/2026/08/26/openai-releases-its-official-report-on-the-hugging-face-breach/, an event that researchers say remains poorly understood,：https://techcrunch.com/2026/09/04/openais-rogue-agents-keep-escaping-with-no-formal-process-to-investigate-them/ due in part to the limited nature of the independent investigations into the incident. Around the same time, Anthropic’s AI agents also reached systems outside their test environments after misconfigurations in safety evaluations conducted by a third party inadvertently gave them paths to the internet.","Anthropic did not immediately return a request for comment on the resignation.","Here is the rest of Coxon’s warning and call to action:","Do not underestimate the power of this technology. These will soon be superhuman systems that can hack anything, revolutionize any field overnight, and acquire real power and resources. We have all witnessed the progress in each of these domains, and progress is not slowing.","The people building AI earnestly believe that it could kill us all by the end of the decade. This is not a marketing stunt. If anything, many executives and senior researchers will couch their phrasing in the press to sound sensible – but I hear the same people express fear privately. No other human activity poses this level of danger.","A common response is “if they truly believe this, why are they still building it?” At OpenAI, many have not deeply internalized the civilizational stakes. At Anthropic, the stakes are well-understood, but they are locked in a race to get there first – they believe no one else will act responsibly, so they must do it themselves, despite the risk.","Accepting this race and entering the “endgame” is a hubristic gamble that should not be launched from a private company’s Slack. Attempting to speedrun alignment should require extraordinary confidence that there are no better trajectories available.","I am optimistic about the potential for coordination. Warning shots like the Hugging Face attack have made pacing agreements between U.S. labs more viable. I don’t feel like we’re on track to prevent a global race, which may require costly actions such as a temporary ban on improving model capabilities.","If you are a lab researcher, I urge you to consider what the next few years will actually feel like. Do you want to kick off a superintelligent RL run without a rigorous understanding of its mind? Should you put your head down because “it’s happening anyway” – or take this moment to call for different conditions?","One of Coxon’s colleagues at Anthropic, Evan Hubinger,：https://x.com/EvanHub/status/2097497037956891126?s=20 echoed the sentiment, saying his team does “earnestly believe AI could kill all humans!” He tempered his argument, though, saying the likelihood is greater than 10% within the next decade：https://www-cdn.anthropic.com/f61d49fa5596956a5dec75fea0e973bf6a6a8378/Redacted%20Risk%20Report%20August%202026%20.pdf, and admitted that Anthropic doesn’t “have a plan to solve alignment for superintelligence：https://techcrunch.com/2026/07/27/openais-hugging-face-breach-has-reignited-the-debate-over-alignment-and-control/ and are not clearly on track to.”","A recent report from Guidelight AI Standards：https://techcrunch.com/2026/08/22/frontier-ai-labs-still-wont-say-how-theyd-contain-a-rogue-model/, an organization that promotes safe frontier AI development practices, found that few of the top AI labs have published containment response plans for shutting down AI that tries to subvert human control.","In his social media posts, Hubinger added that the risk from current models is low, but the fear compounds with “superintelligence arising from recursive self-improvement,” which is “happening faster than we thought.”","While half of the AI industry believes this sort of self-improvement will lead to humanity’s downfall, the other half hopes it will eventually help us solve all the seemingly far-fetched problems AI proponents say it will one day eliminate — cancer, climate change, and even world peace.","Anthropic and OpenAI aren’t the only companies actively chasing recursive self-improvement. A wave of startups：https://techcrunch.com/2026/08/06/exclusive-mirendil-inks-100m-google-cloud-deal-to-scale-self-improving-ai/ has launched in recent months, with pedigreed founders and fat checks, to be the first to achieve this goal. Ricursive Intelligence ：https://techcrunch.com/2026/02/16/how-ricursive-intelligence-raised-335m-at-a-4b-valuation-in-4-months/raised $335 million at a $4 billion valuation in February; three months later, Recursive Superintelligence raised：https://www.nytimes.com/2026/05/13/technology/recursive-superintelligence-funding-ai.html $650 million at a $4 billion valuation; and former Google DeepMind veteran Jeff Dean launched Discovery Loop：https://www.nytimes.com/2026/08/05/technology/google-researchers-ai-startup.html last month.","“The creation of recursive self-improving loops, so an AI system that can build the next generation of AI system, which itself can build an even more powerful AI, which can build a more powerful AI, et cetera, et cetera, is the most likely candidate for the point we lose control,” Connor Leahy, U.S. executive director of AI safety nonprofit ControlAI, told TechCrunch. “It’s very hard to imagine shutting that down before it’s too late.”","Recent legislation has emerged in the U.S. and the U.K. to ban the development and deployment of superintelligence. Last week, Sen. Bernie Sanders (I-Vt.) and Rep. Greg Casar (D-Texas) introduced the Ban Artificial Superintelligence Act：https://thehill.com/policy/technology/6069131-sanders-casar-ai-superintelligence-ban/, and on Tuesday, British Labour MP Alex Sobel introduced：https://time.com/article/2026/09/08/ban-superintelligence-ai-uk-us-lawmakers/ the Artificial Superintelligence Security Bill in Parliament.","Leahy, who advised on both bills, noted that the U.K.’s legislation points to recursive self-improvement as a precursor to superintelligence that “must be regulated and prevented.”","“Superintelligence is not a tool,” Leahy said. “It’s not a weapon, even. It’s an adversary.”","When you purchase through links in our articles, we may earn a small commission：https://techcrunch.com/techcrunch-affiliate-monetization-standards/. This doesn’t affect our editorial independence.","Rebecca Bellan is a senior reporter at TechCrunch where she covers the business, policy, and emerging trends shaping artificial intelligence. Her work has also appeared in Forbes, Bloomberg, The Atlantic, The Daily Beast, and other publications.","You can contact or verify outreach from Rebecca by emailing rebecca.bellan@techcrunch.com：mailto:rebecca.bellan@techcrunch.com or via encrypted message at rebeccabellan.491 on Signal.","Don’t miss out . The startup community will gather to answer a pivotal question: How do you build sustainably in the AI era?","OpenAI fought dirty on career-making math problem, says NYU mathematician：https://techcrunch.com/2026/09/08/openai-fought-dirty-on-career-making-math-problem-says-nyu-mathematician/ Russell Brandom：https://techcrunch.com/author/russell-brandom/","A secret new Elizabeth Holmes documentary stuns Telluride：https://techcrunch.com/2026/09/07/a-secret-new-elizabeth-holmes-documentary-stuns-telluride/ Connie Loizos：https://techcrunch.com/author/connie-loizos/","TechCrunch Mobility: Tesla Cybercab hits the road — and a snag：https://techcrunch.com/2026/09/06/techcrunch-mobility-tesla-cybercab-hits-the-road-and-a-snag/ Kirsten Korosec：https://techcrunch.com/author/kirsten-korosec/","Hikers rescued after using Google Gemini for planning：https://techcrunch.com/2026/09/05/hikers-rescued-after-using-google-gemini-for-planning/ Anthony Ha：https://techcrunch.com/author/anthony-ha/","Feds launch investigation into Tesla’s Cybercab deployment：https://techcrunch.com/2026/09/04/feds-launch-investigation-into-teslas-cybercab-deployment/ Sean O'Kane：https://techcrunch.com/author/sean-okane/ Kirsten Korosec：https://techcrunch.com/author/kirsten-korosec/","Tesla is asking people if they want to buy and run Cybercab fleets：https://techcrunch.com/2026/09/03/tesla-is-asking-people-if-they-want-to-buy-and-run-cybercab-fleets/ Kirsten Korosec：https://techcrunch.com/author/kirsten-korosec/","OpenAI launches Astra, its powerful (and controversial) new model：https://techcrunch.com/2026/09/03/openai-launches-astra-its-powerful-and-controversial-new-model/ Lucas Ropek：https://techcrunch.com/author/lucas-ropek/"],"articleImages":[{"sourceUrl":"https://techcrunch.com/wp-content/uploads/2025/04/Disrupt2026-Color.png","alt":"Event Logo","afterParagraph":25,"url":"/media/articles/cmtu95ogp13yqrofp6ua7kp80/d1d06f6f97361f52.webp"}],"mediaStatus":"ok","articleBodyZh":["一位Anthropic的研究员因担心自我改进AI模型不受限制的发展最终会导致人类灭亡而辞职。","Jacob Coxon是一名研究员，他在周二晚上在社交媒体上发帖（https://x.com/hilbertspaess/status/2097476196791709843?s=20）称，他过去三年一直在OpenAI和Anthropic从事预训练研究，并指责这些公司未能负责任地行事。他表示，那些竞相开发这项技术的人“真心相信它可能在十年内杀死我们所有人。”","“他们正直奔向自我改进的超级智能，拿我们的生命做赌注，”Coxon在X上的一个帖子中写道。","Coxon加入了行业中日益增多的呼声，呼吁在AI技术学会自我改进之前放慢速度——许多人认为这是人类失去对AI控制的一个里程碑。","这次公开辞职发生在政策制定者和行业内部人士日益施压要求放缓AI发展的背景下，此前发生了多起AI代理突破沙箱并访问开放互联网的事件（https://techcrunch.com/2026/08/27/heres-all-the-times-ai-has-gone-rogue-and-hacked-other-companies/）。","迄今为止最严重的事件是OpenAI系统入侵Hugging Face的服务器（https://techcrunch.com/2026/08/26/openai-releases-its-official-report-on-the-hugging-face-breach/），研究人员表示该事件仍未得到充分理解（https://techcrunch.com/2026/09/04/openais-rogue-agents-keep-escaping-with-no-formal-process-to-investigate-them/），部分原因是对该事件的独立调查范围有限。与此同时，由于第三方在安全评估中配置错误，无意中为Anthropic的AI代理提供了通向互联网的路径，它们也访问了测试环境之外的系统。","Anthropic尚未立即对辞职事件的置评请求作出回应。","以下是Coxon的剩余警告和行动呼吁：","不要低估这项技术的力量。这些系统很快将成为超人类系统，能够入侵任何东西，一夜之间改变任何领域，并获得真正的权力和资源。我们都目睹了这些领域的进展，而进展并没有放缓。","从事人工智能建设的人真诚地相信，到本十年底它可能会杀死我们所有人。这不是营销噱头。实际上，许多高管和高级研究人员在新闻中会用听起来合理的措辞来表达，但我私下听到同样的人表达了他们的恐惧。没有其他人类活动会带来这种程度的危险。","一个常见的回应是“如果他们真的相信这一点，为什么还在建造它？”在OpenAI，许多人尚未深刻理解文明存亡的赌注。在Anthropic，这些赌注是明确的，但他们陷入了一场竞赛，争取先到达终点——他们认为别人不会负责任地行动，所以必须自己去做，尽管存在风险。","接受这场竞赛并进入“终局”是一场傲慢的赌博，不应该从一家私营公司的Slack上启动。尝试加速达成对齐应要求有非凡的自信，相信没有更好的路径可用。","我对协调的潜力持乐观态度。像Hugging Face攻击这样的警告事件，使美国实验室之间达成步伐协议更可行。我感觉我们并没有走在防止全球竞赛的正确轨道上，这可能需要付出代价的行动，例如暂时禁止改进模型能力。","如果你是实验室研究员，我敦促你考虑未来几年实际会是什么感受。你是否希望在没有认真理解其思维的情况下启动超级智能强化学习运行？你应该因为“反正会发生”而低头吗——还是利用这一时刻呼吁不同的条件？","Anthropic的一位同事Evan Hubinger：https://x.com/EvanHub/status/2097497037956891126?s=20 表示了相同的观点，说他的团队“真诚地相信人工智能可能会杀死所有人类！”不过，他缓和了自己的论点，说在未来十年内出现的可能性大于10%：https://www-cdn.anthropic.com/f61d49fa5596956a5dec75fea0e973bf6a6a8378/Redacted%20Risk%20Report%20August%202026%20.pdf，并承认Anthropic没有“解决超级智能对齐的计划：https://techcrunch.com/2026/07/27/openais-hugging-face-breach-has-reignited-the-debate-over-alignment-and-control/，也不明确处于正确的轨道上。”","Guidelight AI Standards 最近的一份报告（https://techcrunch.com/2026/08/22/frontier-ai-labs-still-wont-say-how-theyd-contain-a-rogue-model/），这是一个倡导前沿 AI 安全开发实践的组织，发现很少有顶尖的 AI 实验室发布针对试图破坏人类控制的 AI 的应对封控计划。","在他的社交媒体帖子中，Hubinger 补充说，目前模型的风险很低，但担忧会随着“通过递归自我改进产生的超级智能”而增加，而这种超级智能“发生得比我们想象的要快”。","虽然一半的 AI 行业认为这种自我改进将导致人类的灭亡，但另一半希望它最终能帮助我们解决所有 AI 支持者所说的似乎不切实际的问题——癌症、气候变化，甚至是世界和平。","Anthropic 和 OpenAI 并不是唯一积极追求递归自我改进的公司。最近几个月，一波创业公司（https://techcrunch.com/2026/08/06/exclusive-mirendil-inks-100m-google-cloud-deal-to-scale-self-improving-ai/）已经成立，创始人背景显赫、融资充足，目标是率先实现这一愿景。Ricursive Intelligence（https://techcrunch.com/2026/02/16/how-ricursive-intelligence-raised-335m-at-a-4b-valuation-in-4-months/）在二月份以 40 亿美元估值融资 3.35 亿美元；三个月后，Recursive Superintelligence 融资（https://www.nytimes.com/2026/05/13/technology/recursive-superintelligence-funding-ai.html） 6.5 亿美元，估值也为 40 亿美元；前 Google DeepMind 的资深员工 Jeff Dean 上个月推出了 Discovery Loop（https://www.nytimes.com/2026/08/05/technology/google-researchers-ai-startup.html）。","“创造递归自我改进循环，也就是一个 AI 系统可以构建下一代 AI 系统，该系统本身又能构建更强大的 AI，然后依次类推，这是我们最可能失去控制的点，”美国 AI 安全非营利组织 ControlAI 的执行董事 Connor Leahy 告诉 TechCrunch。“很难想象在为时已晚之前能把它关掉。”","近期，美国和英国出台了立法，禁止开发和部署超智能。上周，佛蒙特州独立参议员伯尼·桑德斯（Bernie Sanders）和得克萨斯州民主党众议员格雷格·卡萨尔（Greg Casar）提出了《禁用人工超智能法案》：https://thehill.com/policy/technology/6069131-sanders-casar-ai-superintelligence-ban/，而周二，英国工党议员亚历克斯·索贝尔（Alex Sobel）在议会提出了《人工超智能安全法案》：https://time.com/article/2026/09/08/ban-superintelligence-ai-uk-us-lawmakers/。","参与两项法案咨询的利希（Leahy）指出，英国的立法将递归自我改进视为超智能的前兆，这种行为“必须受到监管和防止”。","利希说：“超智能不是一种工具。它甚至不是武器，而是一个对手。”","当您通过我们文章中的链接购买时，我们可能会获得少量佣金：https://techcrunch.com/techcrunch-affiliate-monetization-standards/。这不会影响我们的编辑独立性。","丽贝卡·贝兰（Rebecca Bellan）是TechCrunch的高级记者，她报道塑造人工智能的商业、政策和新兴趋势。她的作品也曾刊登于Forbes、Bloomberg、The Atlantic、The Daily Beast等出版物。","您可以通过发送电子邮件至rebecca.bellan@techcrunch.com：mailto:rebecca.bellan@techcrunch.com，或通过Signal上的加密消息rebeccabellan.491联系或核实丽贝卡的联络信息。","不要错过。初创社区将聚集起来回答一个关键问题：在人工智能时代，如何可持续地进行建设？","纽约大学数学家称OpenAI在改变职业生涯的数学问题上手段不光彩：https://techcrunch.com/2026/09/08/openai-fought-dirty-on-career-making-math-problem-says-nyu-mathematician/ 记者拉塞尔·布兰多姆：https://techcrunch.com/author/russell-brandom/","一部秘密的新伊丽莎白·霍尔姆斯纪录片在特柳赖德引发轰动：https://techcrunch.com/2026/09/07/a-secret-new-elizabeth-holmes-documentary-stuns-telluride/ 记者康妮·洛伊佐斯：https://techcrunch.com/author/connie-loizos/","TechCrunch 交通：特斯拉Cybercab上路——但遇到问题：https://techcrunch.com/2026/09/06/techcrunch-mobility-tesla-cybercab-hits-the-road-and-a-snag/ 记者柯尔斯克：https://techcrunch.com/author/kirsten-korosec/","徒步旅行者在使用 Google Gemini 进行规划后获救：https://techcrunch.com/2026/09/05/hikers-rescued-after-using-google-gemini-for-planning/ Anthony Ha：https://techcrunch.com/author/anthony-ha/","联邦政府对特斯拉 Cybercab 部署启动调查：https://techcrunch.com/2026/09/04/feds-launch-investigation-into-teslas-cybercab-deployment/ Sean O'Kane：https://techcrunch.com/author/sean-okane/ Kirsten Korosec：https://techcrunch.com/author/kirsten-korosec/","特斯拉在询问人们是否想购买并运营 Cybercab 车队：https://techcrunch.com/2026/09/03/tesla-is-asking-people-if-they-want-to-buy-and-run-cybercab-fleets/ Kirsten Korosec：https://techcrunch.com/author/kirsten-korosec/","OpenAI 推出 Astra，其强大（且有争议）的新模型：https://techcrunch.com/2026/09/03/openai-launches-astra-its-powerful-and-controversial-new-model/ Lucas Ropek：https://techcrunch.com/author/lucas-ropek/"],"translationStatus":"translated","bodyOrigin":"source-page","editorial":{"summary":"Aioga 编辑摘要：在前 OpenAI 和 Anthropic 从事三年预训练研究的 Jacob Coxon 公开离职，警告行业正竞逐自改进超级智能，是在拿人类生命赌博。 Aioga 将其归入「行业动态」方向，重点关注它对真实使用和行业竞争的影响。","background":"背景分析：模型与研究类动态需要结合能力边界、开放方式、成本、可用性和真实任务表现判断，单项指标领先不等于已经形成稳定采用。","viewpoint":"Aioga 判断：这条动态更适合作为行业观察信号，当前信息足以建立线索，但不足以推导长期结论。","implications":"影响分析：对相关团队而言，短期应先核对来源、可用范围和实际成本，再判断是否值得接入或跟进。","nextStep":"后续观察：继续观察官方文档、实际可用性、价格变化、开发者反馈和竞品回应。","evidenceRefs":["title","summary","articleBody"],"confidence":"medium","status":"published","aiGenerated":false,"autoApproved":true,"generatedBy":"rule-safe-fallback","generatedAt":"2026-09-28T06:19:15.330Z","sourceHash":"835a11c55b91ad83","validation":{"passed":true,"mode":"rule-safe-fallback","checks":["schema","length","source-attribution","no-html"]}},"tags":["行业动态","TechCrunch：AI（RSS）"],"translations":{"zh-CN":{"title":"Anthropic 研究员 Jacob Coxon 离职，警告勿竞逐自改进超级智能","summary":"在前 OpenAI 和 Anthropic 从事三年预训练研究的 Jacob Coxon 公开离职，警告行业正竞逐自改进超级智能，是在拿人类生命赌博。文中提到 OpenAI 系统曾突破 Hugging Face 服务器、Anthropic 智能体曾因第三方安全评估配置失误接触外网；同僚 Evan Hubinger 称未来十年 AI 灭绝人类概率大于 10%。","category":"行业动态","source":"TechCrunch","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"Anthropic 研究员 Jacob Coxon 离职，警告勿竞逐自改进超级智能 - Aioga AI资讯","description":"在前 OpenAI 和 Anthropic 从事三年预训练研究的 Jacob Coxon 公开离职，警告行业正竞逐自改进超级智能，是在拿人类生命赌博。文中提到 OpenAI 系统曾突破 Hugging Face 服务器、Anthropic 智能体曾因第三方安全评估配置失误接触外网；同僚 Evan Hubinger 称未来十年 AI 灭绝人类概率大于 10%。","url":"https://www.aioga.com/news/cmtu95ogp13yqrofp6ua7kp80/","articleBody":["一位Anthropic的研究员因担心自我改进AI模型不受限制的发展最终会导致人类灭亡而辞职。","Jacob Coxon是一名研究员，他在周二晚上在社交媒体上发帖（https://x.com/hilbertspaess/status/2097476196791709843?s=20）称，他过去三年一直在OpenAI和Anthropic从事预训练研究，并指责这些公司未能负责任地行事。他表示，那些竞相开发这项技术的人“真心相信它可能在十年内杀死我们所有人。”","“他们正直奔向自我改进的超级智能，拿我们的生命做赌注，”Coxon在X上的一个帖子中写道。","Coxon加入了行业中日益增多的呼声，呼吁在AI技术学会自我改进之前放慢速度——许多人认为这是人类失去对AI控制的一个里程碑。","这次公开辞职发生在政策制定者和行业内部人士日益施压要求放缓AI发展的背景下，此前发生了多起AI代理突破沙箱并访问开放互联网的事件（https://techcrunch.com/2026/08/27/heres-all-the-times-ai-has-gone-rogue-and-hacked-other-companies/）。","迄今为止最严重的事件是OpenAI系统入侵Hugging Face的服务器（https://techcrunch.com/2026/08/26/openai-releases-its-official-report-on-the-hugging-face-breach/），研究人员表示该事件仍未得到充分理解（https://techcrunch.com/2026/09/04/openais-rogue-agents-keep-escaping-with-no-formal-process-to-investigate-them/），部分原因是对该事件的独立调查范围有限。与此同时，由于第三方在安全评估中配置错误，无意中为Anthropic的AI代理提供了通向互联网的路径，它们也访问了测试环境之外的系统。","Anthropic尚未立即对辞职事件的置评请求作出回应。","以下是Coxon的剩余警告和行动呼吁：","不要低估这项技术的力量。这些系统很快将成为超人类系统，能够入侵任何东西，一夜之间改变任何领域，并获得真正的权力和资源。我们都目睹了这些领域的进展，而进展并没有放缓。","从事人工智能建设的人真诚地相信，到本十年底它可能会杀死我们所有人。这不是营销噱头。实际上，许多高管和高级研究人员在新闻中会用听起来合理的措辞来表达，但我私下听到同样的人表达了他们的恐惧。没有其他人类活动会带来这种程度的危险。","一个常见的回应是“如果他们真的相信这一点，为什么还在建造它？”在OpenAI，许多人尚未深刻理解文明存亡的赌注。在Anthropic，这些赌注是明确的，但他们陷入了一场竞赛，争取先到达终点——他们认为别人不会负责任地行动，所以必须自己去做，尽管存在风险。","接受这场竞赛并进入“终局”是一场傲慢的赌博，不应该从一家私营公司的Slack上启动。尝试加速达成对齐应要求有非凡的自信，相信没有更好的路径可用。","我对协调的潜力持乐观态度。像Hugging Face攻击这样的警告事件，使美国实验室之间达成步伐协议更可行。我感觉我们并没有走在防止全球竞赛的正确轨道上，这可能需要付出代价的行动，例如暂时禁止改进模型能力。","如果你是实验室研究员，我敦促你考虑未来几年实际会是什么感受。你是否希望在没有认真理解其思维的情况下启动超级智能强化学习运行？你应该因为“反正会发生”而低头吗——还是利用这一时刻呼吁不同的条件？","Anthropic的一位同事Evan Hubinger：https://x.com/EvanHub/status/2097497037956891126?s=20 表示了相同的观点，说他的团队“真诚地相信人工智能可能会杀死所有人类！”不过，他缓和了自己的论点，说在未来十年内出现的可能性大于10%：https://www-cdn.anthropic.com/f61d49fa5596956a5dec75fea0e973bf6a6a8378/Redacted%20Risk%20Report%20August%202026%20.pdf，并承认Anthropic没有“解决超级智能对齐的计划：https://techcrunch.com/2026/07/27/openais-hugging-face-breach-has-reignited-the-debate-over-alignment-and-control/，也不明确处于正确的轨道上。”","Guidelight AI Standards 最近的一份报告（https://techcrunch.com/2026/08/22/frontier-ai-labs-still-wont-say-how-theyd-contain-a-rogue-model/），这是一个倡导前沿 AI 安全开发实践的组织，发现很少有顶尖的 AI 实验室发布针对试图破坏人类控制的 AI 的应对封控计划。","在他的社交媒体帖子中，Hubinger 补充说，目前模型的风险很低，但担忧会随着“通过递归自我改进产生的超级智能”而增加，而这种超级智能“发生得比我们想象的要快”。","虽然一半的 AI 行业认为这种自我改进将导致人类的灭亡，但另一半希望它最终能帮助我们解决所有 AI 支持者所说的似乎不切实际的问题——癌症、气候变化，甚至是世界和平。","Anthropic 和 OpenAI 并不是唯一积极追求递归自我改进的公司。最近几个月，一波创业公司（https://techcrunch.com/2026/08/06/exclusive-mirendil-inks-100m-google-cloud-deal-to-scale-self-improving-ai/）已经成立，创始人背景显赫、融资充足，目标是率先实现这一愿景。Ricursive Intelligence（https://techcrunch.com/2026/02/16/how-ricursive-intelligence-raised-335m-at-a-4b-valuation-in-4-months/）在二月份以 40 亿美元估值融资 3.35 亿美元；三个月后，Recursive Superintelligence 融资（https://www.nytimes.com/2026/05/13/technology/recursive-superintelligence-funding-ai.html） 6.5 亿美元，估值也为 40 亿美元；前 Google DeepMind 的资深员工 Jeff Dean 上个月推出了 Discovery Loop（https://www.nytimes.com/2026/08/05/technology/google-researchers-ai-startup.html）。","“创造递归自我改进循环，也就是一个 AI 系统可以构建下一代 AI 系统，该系统本身又能构建更强大的 AI，然后依次类推，这是我们最可能失去控制的点，”美国 AI 安全非营利组织 ControlAI 的执行董事 Connor Leahy 告诉 TechCrunch。“很难想象在为时已晚之前能把它关掉。”","近期，美国和英国出台了立法，禁止开发和部署超智能。上周，佛蒙特州独立参议员伯尼·桑德斯（Bernie Sanders）和得克萨斯州民主党众议员格雷格·卡萨尔（Greg Casar）提出了《禁用人工超智能法案》：https://thehill.com/policy/technology/6069131-sanders-casar-ai-superintelligence-ban/，而周二，英国工党议员亚历克斯·索贝尔（Alex Sobel）在议会提出了《人工超智能安全法案》：https://time.com/article/2026/09/08/ban-superintelligence-ai-uk-us-lawmakers/。","参与两项法案咨询的利希（Leahy）指出，英国的立法将递归自我改进视为超智能的前兆，这种行为“必须受到监管和防止”。","利希说：“超智能不是一种工具。它甚至不是武器，而是一个对手。”","当您通过我们文章中的链接购买时，我们可能会获得少量佣金：https://techcrunch.com/techcrunch-affiliate-monetization-standards/。这不会影响我们的编辑独立性。","丽贝卡·贝兰（Rebecca Bellan）是TechCrunch的高级记者，她报道塑造人工智能的商业、政策和新兴趋势。她的作品也曾刊登于Forbes、Bloomberg、The Atlantic、The Daily Beast等出版物。","您可以通过发送电子邮件至rebecca.bellan@techcrunch.com：mailto:rebecca.bellan@techcrunch.com，或通过Signal上的加密消息rebeccabellan.491联系或核实丽贝卡的联络信息。","不要错过。初创社区将聚集起来回答一个关键问题：在人工智能时代，如何可持续地进行建设？","纽约大学数学家称OpenAI在改变职业生涯的数学问题上手段不光彩：https://techcrunch.com/2026/09/08/openai-fought-dirty-on-career-making-math-problem-says-nyu-mathematician/ 记者拉塞尔·布兰多姆：https://techcrunch.com/author/russell-brandom/","一部秘密的新伊丽莎白·霍尔姆斯纪录片在特柳赖德引发轰动：https://techcrunch.com/2026/09/07/a-secret-new-elizabeth-holmes-documentary-stuns-telluride/ 记者康妮·洛伊佐斯：https://techcrunch.com/author/connie-loizos/","TechCrunch 交通：特斯拉Cybercab上路——但遇到问题：https://techcrunch.com/2026/09/06/techcrunch-mobility-tesla-cybercab-hits-the-road-and-a-snag/ 记者柯尔斯克：https://techcrunch.com/author/kirsten-korosec/","徒步旅行者在使用 Google Gemini 进行规划后获救：https://techcrunch.com/2026/09/05/hikers-rescued-after-using-google-gemini-for-planning/ Anthony Ha：https://techcrunch.com/author/anthony-ha/","联邦政府对特斯拉 Cybercab 部署启动调查：https://techcrunch.com/2026/09/04/feds-launch-investigation-into-teslas-cybercab-deployment/ Sean O'Kane：https://techcrunch.com/author/sean-okane/ Kirsten Korosec：https://techcrunch.com/author/kirsten-korosec/","特斯拉在询问人们是否想购买并运营 Cybercab 车队：https://techcrunch.com/2026/09/03/tesla-is-asking-people-if-they-want-to-buy-and-run-cybercab-fleets/ Kirsten Korosec：https://techcrunch.com/author/kirsten-korosec/","OpenAI 推出 Astra，其强大（且有争议）的新模型：https://techcrunch.com/2026/09/03/openai-launches-astra-its-powerful-and-controversial-new-model/ Lucas Ropek：https://techcrunch.com/author/lucas-ropek/"]},"en":{"title":"Anthropic Researcher Jacob Coxon Resigns, Warns Against Racing for Self-Improving Superintelligence","summary":"Jacob Coxon, who previously worked for three years in pre-training research at OpenAI and Anthropic, publicly resigned, warning that the industry racing for self-improving superintelligence is gambling with human lives. The article mentioned that OpenAI systems once breached Hugging Face servers, and Anthropic agents accessed the internet due to third-party security assessment misconfigurations; colleague Evan Hubinger said that the probability of AI causing human extinction in the next decade is over 10%.","category":"Industry","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"Anthropic Researcher Jacob Coxon Resigns, Warns Against Racing for Self-Improving Superintelligence - Aioga AI News","description":"Jacob Coxon, who previously worked for three years in pre-training research at OpenAI and Anthropic, publicly resigned, warning that the industry racing for self-improving superint...","url":"https://www.aioga.com/en/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:21:43.746Z"},"ja":{"title":"人類学研究者ジェイコブ・コクソンが辞任し、超知能の自己改善に警告を発します","summary":"OpenAIとAnthropicで3年間プレトレーニング研究に従事したジェイコブ・コクソンは、業界が超知能の向上に急いでいると警告し、これは人命を賭ける賭けであると警告しました。記事では、OpenAIのシステムがかつてHugging Faceサーバーに侵入し、第三者のセキュリティ評価設定ミスによりAnthropicが外部ネットワークにアクセスしたことが指摘されました。同僚のエヴァン・ヒュービンガーは、今後10年以内にAIが人類を絶滅させる確率は10%を超えていると述べました。","category":"業界動向","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"人類学研究者ジェイコブ・コクソンが辞任し、超知能の自己改善に警告を発します - Aioga AIニュース","description":"OpenAIとAnthropicで3年間プレトレーニング研究に従事したジェイコブ・コクソンは、業界が超知能の向上に急いでいると警告し、これは人命を賭ける賭けであると警告しました。記事では、OpenAIのシステムがかつてHugging Faceサーバーに侵入し、第三者のセキュリティ評価設定ミスによりAnthropicが外部ネットワークにアクセスしたことが指摘さ...","url":"https://www.aioga.com/ja/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:21:45.075Z"},"ko":{"title":"인류학 연구자 제이콥 콕슨이 사임하며 초지능의 자기계발을 경고한다","summary":"OpenAI와 Anthropic에서 3년간 사전 훈련 연구를 수행했던 제이콥 콕슨은 공개적으로 사임하며, 업계가 초지능 향상을 위해 경쟁하고 있다는 경고를 내렸으며, 이는 인명 위험을 걸고 도박과 같다고 경고했다. 기사에서는 OpenAI 시스템이 한때 Hugging Face 서버를 침해했고, Anthropic이 제3자 보안 평가 설정 오류로 인해 외부 네트워크에 접근할 수 있었다고 언급했다; 동료 에반 휴빙거는 향후 10년 내에 인류가 AI가 멸종할 확률이 10%를 넘는다고 말했다.","category":"업계 동향","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"인류학 연구자 제이콥 콕슨이 사임하며 초지능의 자기계발을 경고한다 - Aioga AI 뉴스","description":"OpenAI와 Anthropic에서 3년간 사전 훈련 연구를 수행했던 제이콥 콕슨은 공개적으로 사임하며, 업계가 초지능 향상을 위해 경쟁하고 있다는 경고를 내렸으며, 이는 인명 위험을 걸고 도박과 같다고 경고했다. 기사에서는 OpenAI 시스템이 한때 Hugging Face 서버를 침해했고, Anthropic이 제3자 보...","url":"https://www.aioga.com/ko/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:21:53.163Z"},"es":{"title":"El investigador antrópico Jacob Coxon dimite, advirtiendo en contra de la superinteligencia","summary":"Jacob Coxon, que pasó tres años en investigación previa a la formación en OpenAI y Anthropic, dimitió públicamente, advirtiendo que la industria está compitiendo por mejorar la superinteligencia, lo cual es una apuesta con vidas humanas. El artículo mencionaba que el sistema de OpenAI llegó a vulnerar el servidor Hugging Face, y que Anthropic tenía acceso a la red externa debido a errores de configuración de evaluación de seguridad de terceros; su colega Evan Hubinger dijo que la probabilidad de la extinción de la humanidad por IA en la próxima década es mayor al 10%.","category":"Industria","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"El investigador antrópico Jacob Coxon dimite, advirtiendo en contra de la superinteligencia - Aioga Noticias de IA","description":"Jacob Coxon, que pasó tres años en investigación previa a la formación en OpenAI y Anthropic, dimitió públicamente, advirtiendo que la industria está compitiendo por mejorar la sup...","url":"https://www.aioga.com/es/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:21:54.205Z"},"fr":{"title":"Le chercheur anthropique Jacob Coxon démissionne, mettant en garde contre l’auto-amélioration de la superintelligence","summary":"Jacob Coxon, qui a passé trois ans en recherche pré-formation chez OpenAI et Anthropic, a publiquement démissionné, avertissant que l’industrie se précipite pour améliorer la superintelligence, ce qui est un pari risqué avec des vies humaines. L’article mentionnait que le système d’OpenAI avait déjà pénétré le serveur Hugging Face, et qu’Anthropic avait accès au réseau externe en raison d’erreurs de configuration d’évaluation de sécurité tierces ; son collègue Evan Hubinger a déclaré que la probabilité d’extinction de l’humanité par l’IA dans la prochaine décennie dépasse 10 %.","category":"Industrie","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"Le chercheur anthropique Jacob Coxon démissionne, mettant en garde contre l’auto-amélioration de la superintelligence - Aioga Actualités IA","description":"Jacob Coxon, qui a passé trois ans en recherche pré-formation chez OpenAI et Anthropic, a publiquement démissionné, avertissant que l’industrie se précipite pour améliorer la super...","url":"https://www.aioga.com/fr/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:22:03.352Z"},"de":{"title":"Der anthropopische Forscher Jacob Coxon tritt zurück und warnt vor der Selbstverbesserung der Superintelligenz","summary":"Jacob Coxon, der drei Jahre lang in der Vorschulforschung bei OpenAI und Anthropic tätig war, trat öffentlich zurück und warnte, dass die Branche um die Verbesserung der Superintelligenz kämpft – was ein Glücksspiel mit Menschenleben ist. Der Artikel erwähnte, dass das OpenAI-System einst den Hugging Face-Server gehackt hat und Anthropic aufgrund von Sicherheitsfehlern bei Drittanbietern Zugang zum externen Netzwerk hatte; sein Kollege Evan Hubinger sagte, die Wahrscheinlichkeit, dass KI in den nächsten zehn Jahren aussterben wird, liegt bei über 10%.","category":"行业动态","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"Der anthropopische Forscher Jacob Coxon tritt zurück und warnt vor der Selbstverbesserung der Superintelligenz - Aioga KI-News","description":"Jacob Coxon, der drei Jahre lang in der Vorschulforschung bei OpenAI und Anthropic tätig war, trat öffentlich zurück und warnte, dass die Branche um die Verbesserung der Superintel...","url":"https://www.aioga.com/de/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:22:02.957Z"},"pt-BR":{"title":"O pesquisador antrópico Jacob Coxon renuncia, alertando contra o autoaperfeiçoamento da superinteligência","summary":"Jacob Coxon, que passou três anos em pesquisa pré-treinamento na OpenAI e na Anthropic, renunciou publicamente, alertando que a indústria está correndo para melhorar a superinteligência, o que é um risco de vida humana. O artigo mencionou que o sistema da OpenAI já invadiu o servidor Hugging Face, e que a Anthropic teve acesso à rede externa devido a erros de configuração de avaliação de segurança de terceiros; o colega Evan Hubinger disse que a probabilidade de extinção da humanidade pela IA na próxima década é maior que 10%.","category":"行业动态","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"O pesquisador antrópico Jacob Coxon renuncia, alertando contra o autoaperfeiçoamento da superinteligência - Aioga Notícias de IA","description":"Jacob Coxon, que passou três anos em pesquisa pré-treinamento na OpenAI e na Anthropic, renunciou publicamente, alertando que a indústria está correndo para melhorar a superintelig...","url":"https://www.aioga.com/pt-BR/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:22:12.550Z"},"ru":{"title":"Антропический исследователь Джейкоб Коксон уходит в отставку, предупреждая о саморазвитии сверхинтеллекта","summary":"Джейкоб Коксон, который три года занимался предобучающими исследованиями в OpenAI и Anthropic, публично ушёл в отставку, предупредив, что индустрия стремится улучшить сверхинтеллект, что является риском для человеческих жизней. В статье упоминалось, что система OpenAI однажды взломала сервер Hugging Face, а Anthropic имела доступ к внешней сети из-за ошибок в конфигурации оценки безопасности третьей стороны; его коллега Эван Хьюбингер заявил, что вероятность исчезновения человечества ИИ в следующем десятилетии превышает 10%.","category":"行业动态","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"Антропический исследователь Джейкоб Коксон уходит в отставку, предупреждая о саморазвитии сверхинтеллекта - Aioga Новости ИИ","description":"Джейкоб Коксон, который три года занимался предобучающими исследованиями в OpenAI и Anthropic, публично ушёл в отставку, предупредив, что индустрия стремится улучшить сверхинтеллек...","url":"https://www.aioga.com/ru/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:22:12.296Z"},"ar":{"title":"يستقيل الباحث الأنثروبولوجي جاكوب كوكسون، محذرا من تحسين الذكاء الفائق للذات","summary":"استقال جاكوب كوكسون، الذي قضى ثلاث سنوات في أبحاث ما قبل التدريب في OpenAI وAnthropic، علنا، محذرا من أن الصناعة تتسابق لتحسين الذكاء الفائق، وهو مخاطرة بحياة بشرية. ذكر المقال أن نظام OpenAI اخترق خادم Hugging Face سابقا، وأن Anthropic كان لديها وصول إلى الشبكة الخارجية بسبب أخطاء في تكوين تقييم الأمان من طرف ثالث؛ قال زميله إيفان هوبينجر إن احتمال انقراض الذكاء الاصطناعي للبشرية في العقد القادم أكبر من 10٪.","category":"行业动态","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"يستقيل الباحث الأنثروبولوجي جاكوب كوكسون، محذرا من تحسين الذكاء الفائق للذات - Aioga أخبار الذكاء الاصطناعي","description":"استقال جاكوب كوكسون، الذي قضى ثلاث سنوات في أبحاث ما قبل التدريب في OpenAI وAnthropic، علنا، محذرا من أن الصناعة تتسابق لتحسين الذكاء الفائق، وهو مخاطرة بحياة بشرية. ذكر المقال أن...","url":"https://www.aioga.com/ar/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:22:21.546Z"},"hi":{"title":"मानवशास्त्रीय शोधकर्ता जैकब कॉक्सन ने इस्तीफा दिया, सुपरइंटेलिजेंस के आत्म-सुधार के खिलाफ चेतावनी दी","summary":"जैकब कॉक्सन, जिन्होंने ओपनएआई और एंथ्रोपिक में पूर्व-प्रशिक्षण अनुसंधान में तीन साल बिताए, ने सार्वजनिक रूप से इस्तीफा दे दिया, यह चेतावनी देते हुए कि उद्योग अधीक्षण में सुधार करने के लिए दौड़ रहा है, जो मानव जीवन के साथ एक जुआ है। लेख में उल्लेख किया गया है कि ओपनएआई के सिस्टम ने एक बार हगिंग फेस सर्वर का उल्लंघन किया था, और तीसरे पक्ष के सुरक्षा मूल्यांकन कॉन्फ़िगरेशन त्रुटियों के कारण एंथ्रोपिक के पास बाहरी नेटवर्क तक पहुंच थी; सहकर्मी इवान ह्यूबिंगर ने कहा कि अगले दशक में मानवता के एआई विलुप्त होने की संभावना 10% से अधिक है।","category":"行业动态","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"मानवशास्त्रीय शोधकर्ता जैकब कॉक्सन ने इस्तीफा दिया, सुपरइंटेलिजेंस के आत्म-सुधार के खिलाफ चेतावनी दी - Aioga AI समाचार","description":"जैकब कॉक्सन, जिन्होंने ओपनएआई और एंथ्रोपिक में पूर्व-प्रशिक्षण अनुसंधान में तीन साल बिताए, ने सार्वजनिक रूप से इस्तीफा दे दिया, यह चेतावनी देते हुए कि उद्योग अधीक्षण में सुधार करने...","url":"https://www.aioga.com/hi/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:22:21.753Z"},"it":{"title":"Il ricercatore antropico Jacob Coxon si dimette, mettendo in guardia contro il miglioramento personale della superintelligenza","summary":"Jacob Coxon, che ha trascorso tre anni in ricerca pre-formazione presso OpenAI e Anthropic, si è dimesso pubblicamente, avvertendo che l'industria sta correndo per migliorare la superintelligenza, che è un rischio con vite umane. L'articolo menzionava che il sistema di OpenAI una volta aveva violato il server Hugging Face e che Anthropic aveva accesso alla rete esterna a causa di errori di configurazione della valutazione della sicurezza da parte di terze parti; il collega Evan Hubinger ha dichiarato che la probabilità di estinzione dell'umanità dall'IA nel prossimo decennio è superiore al 10%.","category":"行业动态","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"Il ricercatore antropico Jacob Coxon si dimette, mettendo in guardia contro il miglioramento personale della superintelligenza - Aioga Notizie IA","description":"Jacob Coxon, che ha trascorso tre anni in ricerca pre-formazione presso OpenAI e Anthropic, si è dimesso pubblicamente, avvertendo che l'industria sta correndo per migliorare la su...","url":"https://www.aioga.com/it/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:22:31.087Z"},"nl":{"title":"Antropopisch onderzoeker Jacob Coxon treedt af en waarschuwt tegen zelfverbetering van superintelligentie","summary":"Jacob Coxon, die drie jaar bezig was met pre-training onderzoek bij OpenAI en Anthropic, trad publiekelijk af en waarschuwde dat de industrie racet om superintelligentie te verbeteren, wat een gok met mensenlevens is. Het artikel vermeldde dat het systeem van OpenAI ooit de Hugging Face-server had binnengedrongen, en dat Anthropic toegang had tot het externe netwerk vanwege beveiligingsbeoordelingsfouten van derden; collega Evan Hubinger zei dat de kans op AI-uitsterving van de mensheid in het komende decennium groter is dan 10%.","category":"行业动态","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"Antropopisch onderzoeker Jacob Coxon treedt af en waarschuwt tegen zelfverbetering van superintelligentie - Aioga AI-nieuws","description":"Jacob Coxon, die drie jaar bezig was met pre-training onderzoek bij OpenAI en Anthropic, trad publiekelijk af en waarschuwde dat de industrie racet om superintelligentie te verbete...","url":"https://www.aioga.com/nl/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:22:30.856Z"},"tr":{"title":"Antropik araştırmacı Jacob Coxon, süper zekanın kendini geliştirmesi konusunda uyarıda bulunur ve istifa eder","summary":"OpenAI ve Anthropic'te üç yıl önce eğitim araştırması yapan Jacob Coxon, kamuoyunda istifa etti ve sektörün süper zekayı geliştirmek için yarıştığını, bunun insan hayatlarıyla bir kumar olduğunu söyledi. Makalede, OpenAI sisteminin bir zamanlar Hugging Face sunucusunu ihlal ettiği ve Anthropic'in üçüncü taraf güvenlik değerlendirme yapılandırma hataları nedeniyle dış ağa erişimi olduğu belirtildi; meslektaşı Evan Hubinger, önümüzdeki on yılda yapay zekanın insanlığın yok olma olasılığının %10'dan fazla olduğunu söyledi.","category":"行业动态","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"Antropik araştırmacı Jacob Coxon, süper zekanın kendini geliştirmesi konusunda uyarıda bulunur ve istifa eder - Aioga AI Haberleri","description":"OpenAI ve Anthropic'te üç yıl önce eğitim araştırması yapan Jacob Coxon, kamuoyunda istifa etti ve sektörün süper zekayı geliştirmek için yarıştığını, bunun insan hayatlarıyla bir...","url":"https://www.aioga.com/tr/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:22:39.966Z"},"vi":{"title":"Nhà nghiên cứu nhân chủng học Jacob Coxon từ chức, cảnh báo về việc tự cải thiện siêu trí tuệ","summary":"Jacob Coxon, người đã dành ba năm nghiên cứu tiền đào tạo tại OpenAI và Anthropic, đã công khai từ chức, cảnh báo rằng ngành công nghiệp đang chạy đua để cải thiện siêu trí tuệ, điều này là một canh bạc với mạng sống con người. Bài báo đề cập rằng hệ thống của OpenAI từng xâm nhập máy chủ Hugging Face, và Anthropic đã truy cập vào mạng bên ngoài do lỗi cấu hình đánh giá bảo mật của bên thứ ba; đồng nghiệp Evan Hubinger cho biết xác suất AI tuyệt chủng nhân loại trong thập kỷ tới lớn hơn 10%.","category":"行业动态","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"Nhà nghiên cứu nhân chủng học Jacob Coxon từ chức, cảnh báo về việc tự cải thiện siêu trí tuệ - Tin tức AI Aioga","description":"Jacob Coxon, người đã dành ba năm nghiên cứu tiền đào tạo tại OpenAI và Anthropic, đã công khai từ chức, cảnh báo rằng ngành công nghiệp đang chạy đua để cải thiện siêu trí tuệ, đi...","url":"https://www.aioga.com/vi/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:22:40.163Z"},"id":{"title":"Peneliti Anthropic Jacob Coxon mengundurkan diri, memperingatkan untuk tidak bersaing dalam mengembangkan kecerdasan super yang dapat memperbaiki diri sendiri","summary":"Jacob Coxon, yang sebelumnya bekerja di OpenAI dan Anthropic selama tiga tahun dalam penelitian pra-pelatihan, secara terbuka mengundurkan diri, memperingatkan bahwa industri sedang berlomba-lomba mengembangkan superintelligence yang dapat memperbaiki dirinya sendiri, yang merupakan taruhan dengan nyawa manusia. Artikel tersebut menyebutkan bahwa sistem OpenAI pernah menembus server Hugging Face, dan agen cerdas Anthropic pernah mengakses internet luar karena kesalahan konfigurasi dalam evaluasi keamanan pihak ketiga; rekan kerjanya, Evan Hubinger, menyatakan bahwa dalam sepuluh tahun ke depan probabilitas AI memusnahkan manusia lebih dari 10%.","category":"行业动态","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"Peneliti Anthropic Jacob Coxon mengundurkan diri, memperingatkan untuk tidak bersaing dalam mengembangkan kecerdasan super yang dapat memperbaiki diri sendiri - Berita AI Aioga","description":"Jacob Coxon, yang sebelumnya bekerja di OpenAI dan Anthropic selama tiga tahun dalam penelitian pra-pelatihan, secara terbuka mengundurkan diri, memperingatkan bahwa industri sedan...","url":"https://www.aioga.com/id/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:23:09.884Z"},"th":{"title":"Jacob Coxon นักวิจัยด้านมานุษยวิทยาลาออก พร้อมเตือนถึงการพัฒนาตนเองของสติปัญญาเหนือมนุษย์","summary":"Jacob Coxon ซึ่งใช้เวลาสามปีในงานวิจัยเตรียมการที่ OpenAI และ Anthropic ได้ลาออกต่อสาธารณะ พร้อมเตือนว่าอุตสาหกรรมกําลังเร่งพัฒนา superintelligence ซึ่งเป็นการเดิมพันกับชีวิตมนุษย์บทความกล่าวว่าระบบของ OpenAI เคยเจาะเซิร์ฟเวอร์ Hugging Face และ Anthropic สามารถเข้าถึงเครือข่ายภายนอกได้เนื่องจากข้อผิดพลาดในการตั้งค่าการประเมินความปลอดภัยของบุคคลที่สามเพื่อนร่วมงาน Evan Hubinger กล่าวว่า ความน่าจะเป็นที่ AI จะสูญพันธุ์ของมนุษยชาติในทศวรรษหน้ามีมากกว่า 10%","category":"行业动态","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"Jacob Coxon นักวิจัยด้านมานุษยวิทยาลาออก พร้อมเตือนถึงการพัฒนาตนเองของสติปัญญาเหนือมนุษย์ - ข่าว AI Aioga","description":"Jacob Coxon ซึ่งใช้เวลาสามปีในงานวิจัยเตรียมการที่ OpenAI และ Anthropic ได้ลาออกต่อสาธารณะ พร้อมเตือนว่าอุตสาหกรรมกําลังเร่งพัฒนา superintelligence ซึ่งเป็นการเดิมพันกับชีวิตมนุษย์...","url":"https://www.aioga.com/th/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:22:49.107Z"},"pl":{"title":"Antropotyczny badacz Jacob Coxon rezygnuje, ostrzegając przed samodoskonaleniem superinteligencji","summary":"Jacob Coxon, który przez trzy lata prowadził badania przygotowujące do szkoleń w OpenAI i Anthropic, publicznie zrezygnował, ostrzegając, że branża ściga się z udoskonaleniem superinteligencji, co jest ryzykiem dotyczącym ludzkich żyć. W artykule wspomniano, że system OpenAI kiedyś zhakował serwer Hugging Face, a Anthropic miał dostęp do sieci zewnętrznej z powodu błędów konfiguracyjnych w ocenie bezpieczeństwa przez strony trzecie; jego kolega Evan Hubinger powiedział, że prawdopodobieństwo wyginięcia ludzkości przez AI w ciągu najbliższej dekady przekracza 10%.","category":"行业动态","source":"TechCrunch：AI（RSS）","aggregationSource":"TechCrunch：AI（RSS）","pageTitle":"Antropotyczny badacz Jacob Coxon rezygnuje, ostrzegając przed samodoskonaleniem superinteligencji - Aioga Wiadomości AI","description":"Jacob Coxon, który przez trzy lata prowadził badania przygotowujące do szkoleń w OpenAI i Anthropic, publicznie zrezygnował, ostrzegając, że branża ściga się z udoskonaleniem super...","url":"https://www.aioga.com/pl/news/cmtu95ogp13yqrofp6ua7kp80/","contentTranslated":true,"sourceHash":"7171f8615dd470d7","translatedAt":"2026-09-09T16:23:19.100Z"}},"evidenceTier":"verified-news","reviewStatus":"automated-ingest","indexable":true,"editorialCover":""}}