比尔·盖茨对人工智能可能被滥用深感担忧。这名微软联合创始人在接受美国全国广播公司《Meet the Press》采访时，就这项技术发出严厉警告，称“人工智能的威力绝对足以推动一些事件，最终造成十亿人死亡。”盖茨还表示，该行业需要政府监管。他说：“你需要执法部门和政治人物参与讨论，明确安全保障和监测机制应当采取什么形式。”盖茨的言论与Anthropic首席执行官达里奥·阿莫代伊的观点不谋而合。阿莫代伊曾在9月发表的一篇文章中，呼吁前沿人工智能公司放慢发展步伐，并称“通过监管来控制节奏是最有效的方法”。在周四发布的一段《Meet the Press》预告片段中，盖茨为将于周日播出的完整采访作铺垫时表示，他担心不法分子会得到先进的人工智能工具。Anthropic最近发布的一份报告详细列举了国家和非国家行为体利用Claude策划无人机袭击以及开展危险生物研究的多个案例。阿莫代伊主张控制人工智能发展节奏的文章获得了埃隆·马斯克和萨姆·奥尔特曼的支持。此前，Anthropic前研究员雅各布·科克森曾警告称，人工智能实验室正在“拿我们的生命赌博”，其辞职帖子在网上广为流传。迄今为止，这些有关人工智能安全的警告仍未能说服华盛顿方面暂停或放慢该技术的发展。特朗普总统则嘲笑阿莫代伊，指责他“假装成一个‘完美的小天使’”。特朗普已明确表现出对监管人工智能兴趣不大，不过他上周确实承诺成立一支“人工智能部队”，对人工智能公司进行“监督”。就连微软首席执行官兼董事长萨提亚·纳德拉似乎也不愿主张对这项技术实施全面监管。纳德拉最近亮相《All-In播客》时表示，人工智能公司应自行处理相关问题。
Bill Gates has some serious fears about how AI can be misused. The cofounder of Microsoft made grave warnings about the tech in an interview with NBC's "Meet the Press," saying that "AI is certainly powerful enough to drive events that cause a billion deaths." Gates also said that the industry needs government regulation. "You need law enforcement and the politicians to get into the discussion about what safeguards and monitoring look like," he said. Gates' comments echo those of Anthropic CEO Dario Amodei, who said in a September essay calling on frontier AI companies to slow down development that the "most effective method of pacing is via regulation." In a clip of the "Meet the Press" published on Thursday ahead of the full interview on Sunday, Gates said he worries about bad actors getting their hands on advanced AI tools. Anthropic recently published a report detailing several instances where state and non-state actors used Claude to plan drone attacks and for dangerous biological research. Amodei's essay calling for AI pacing, which received support from Elon Musk and Sam Altman, was published in the wake of warnings from ex-Anthropic researcher Jacob Coxon, who wrote a viral post about resigning and AI labs "gambling with our lives." The AI safety warnings have so far failed to convince Washington that development of the tech should be paused or slowed down. President Donald Trump, for his part, mocked Amodei, accusing him of "pretending to be a 'perfect little angel.'" Trump has signaled strong disinterest in regulating AI, though he did pledge last week to create an "AI Force" that would "watch over" AI companies. Even Satya Nadella, Microsoft's CEO and chairman, has appeared reticent to call for all-out regulation of the tech. In a recent appearance on the All-In Podcast, Nadella said AI companies should handle things on their own. "If you see a show-stopper," Nadella said, referring to anomalies in code that could produce rogue AI agents, "stop the show." Though there is broad agreement across Silicon Valley that AI companies should implement more robust safeguards to protect against incidents like OpenAI's breach of Hugging Face, CEOs differ on how such safeguards should be enforced. This month, Nadella warned in a memo to Microsoft employees viewed by Business Insider that companies must take time to make the technology safe. Gates, who now serves as a technology advisor to Nadella and other Microsoft executives, was previously one of AI's most outspoken optimists before changing course over the summer. In an essay he published in August, Gates said the world has no plan to roll out AI safely. "I believe we need time to prepare for the period of social, political, and economic upheaval we are about to enter," he said. "If someone had a credible plan for slowing down AI advances globally, I would likely support it."
“如果你发现一个致命问题，”纳德拉说道，他指的是代码中可能导致失控AI智能体的异常，“就停止演出。”尽管硅谷普遍认同，AI公司应实施更 robust 的安全保障措施，以防止类似OpenAI入侵Hugging Face的事件发生，但首席执行官们在如何执行这些保障措施上意见不一。本月，纳德拉在一份被《商业内幕》看到的致微软员工的备忘录中警告称，公司必须花时间确保技术的安全性。盖茨目前担任纳德拉及其他微软高管的技术顾问，他此前是AI领域最直言不讳的乐观主义者之一，但在今年夏天改变了立场。在他八月份发表的一篇文章中，盖茨表示，世界没有安全部署AI的计划。“我认为我们需要时间来为即将进入的社会、政治和经济动荡时期做准备，”他说。“如果有人有一个可信的计划来减缓全球AI的进步，我可能会支持它。”
Bill Gates has some serious fears about how AI can be misused. The cofounder of Microsoft made grave warnings about the tech in an interview with NBC's "Meet the Press," saying that "AI is certainly powerful enough to drive events that cause a billion deaths." Gates also said that the industry needs government regulation. "You need law enforcement and the politicians to get into the discussion about what safeguards and monitoring look like," he said. Gates' comments echo those of Anthropic CEO Dario Amodei, who said in a September essay calling on frontier AI companies to slow down development that the "most effective method of pacing is via regulation." In a clip of the "Meet the Press" published on Thursday ahead of the full interview on Sunday, Gates said he worries about bad actors getting their hands on advanced AI tools. Anthropic recently published a report detailing several instances where state and non-state actors used Claude to plan drone attacks and for dangerous biological research. Amodei's essay calling for AI pacing, which received support from Elon Musk and Sam Altman, was published in the wake of warnings from ex-Anthropic researcher Jacob Coxon, who wrote a viral post about resigning and AI labs "gambling with our lives." The AI safety warnings have so far failed to convince Washington that development of the tech should be paused or slowed down. President Donald Trump, for his part, mocked Amodei, accusing him of "pretending to be a 'perfect little angel.'" Trump has signaled strong disinterest in regulating AI, though he did pledge last week to create an "AI Force" that would "watch over" AI companies. Even Satya Nadella, Microsoft's CEO and chairman, has appeared reticent to call for all-out regulation of the tech. In a recent appearance on the All-In Podcast, Nadella said AI companies should handle things on their own. "If you see a show-stopper," Nadella said, referring to anomalies in code that could produce rogue AI agents, "stop the show." Though there is broad agreement across Silicon Valley that AI companies should implement more robust safeguards to protect against incidents like OpenAI's breach of Hugging Face, CEOs differ on how such safeguards should be enforced. This month, Nadella warned in a memo to Microsoft employees viewed by Business Insider that companies must take time to make the technology safe. Gates, who now serves as a technology advisor to Nadella and other Microsoft executives, was previously one of AI's most outspoken optimists before changing course over the summer. In an essay he published in August, Gates said the world has no plan to roll out AI safely. "I believe we need time to prepare for the period of social, political, and economic upheaval we are about to enter," he said. "If someone had a credible plan for slowing down AI advances globally, I would likely support it."