That is a difficult problem ... because they’re open-weight, you can’t really work with the companies to fix their models, because once they release them onto the internet, people just take them and they can change whatever it is they want with those models
这句话暴露了"关停开关"立法思路的结构性盲区——它对闭权重模型有效，但对开放权重模型基本失灵，因为权重一旦发布就脱离了原厂商的控制。这个漏洞恰好和 EP.97 故事线 B 的核心论点相互印证：出口管制/关停机制能管住"中心化可控的东西"，管不住已经扩散出去的模型权重和能力。
We don’t slow down how they build their models. We just say, look, after you complete your model, and it turns out that it might have some sort of really bad catastrophic risk, or some sort of flaw, then you need to have ability to shut it down, or the government has to have ability to shut it down
众议员 Ted Lieu 把这项立法的定位说得非常清楚：不干预训练过程，只要求"事后必须有能力关停"。这正是 EP.97 故事线 A 强调的"基础设施抓手"思路——监管重点从"审查模型该不该被造出来"转移到"确保任何已经存在的模型都有可靠的关停机制"，与汽车碰撞测试的类比也呼应了本期对"evaluation infrastructure"重要性的讨论。
We need to get this bill across the finish line this year because the advanced closed-weight models are already doing, as you noted, unauthorized hacks of other companies
这是 EP.97 故事线 A"监管抓手从发布前审查转向事故披露+基础设施"论点在立法层面的直接证据——AI Kill Switch Act 的推动力不是理论风险，而是 OpenAI/Anthropic/Meta 已经连续披露的真实入侵事件。国会议员用"事故已经在发生"作为立法紧迫性的论据，说明监管话语正在从"防患于未然"转向"响应已发生的失控"。
The letter lands two months after the White House Office of Science and Technology Policy issued a memorandum that pledged to help AI companies detect and coordinate against industrial-scale distillation.
这句话提供了重要的政策背景，表明此事件发生在特定的政策环境下。需要了解该备忘录的具体内容和实施情况，以及它如何影响Anthropic和Alibaba的行为。这涉及到政府政策与科技行业实践之间的互动关系，值得深入了解。
the message is clear: The AI industry isn't immune from U.S. government interference
大多数人可能认为AI技术的前沿性质使其能够规避传统监管框架，但作者认为政府的禁令明确传递了一个信息：即使是尖端AI技术也不能摆脱政府干预。这与科技行业自认为能够自我监管的普遍认知相悖。
The US government's Anthropic models ban was never about an AI jailbreak
大多数人认为政府禁止Anthropic的AI模型是出于安全考虑，特别是担心AI越狱风险，但作者认为这并非真正原因。这是一个非共识观点，挑战了公众对政府监管AI的普遍理解。
If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers.
大多数人认为政府的安全审查是合理的预防措施，但作者认为这种标准如果普遍应用，实际上会停止整个行业的前沿模型部署，这暗示了政府安全标准可能过于严苛，阻碍了AI创新和技术进步。
If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers.
大多数人认为政府对AI模型的安全监管是必要的保护措施，但作者认为如果这种标准（因发现狭窄的潜在越狱就召回商业模型）在整个行业应用，将基本上停止所有前沿模型提供商的新模型部署。这是一个挑战AI监管共识的观点。
Around the world, AI systems are being deployed at scale with remarkably little institutional oversight. There is no AI safety board. The US Federal Trade Commission has jurisdiction over unfair practices but limited authority over algorithmic design. The National Institute of Standards and Technology publishes guidance that most companies ignore. The EU AI Act is partially in force but addresses only a sliver of the deployment surface.
This regulatory landscape summary is unusually blunt for MIT Technology Review: four specific institutions listed, four specific ways each falls short. The cumulative picture is that the entire institutional stack — domestic regulators, international standards bodies, supranational legislation — is structurally inadequate to the speed and scope of AI deployment. This is the governance gap that makes the shareholder argument necessary.
This encyclical doesn't break new ground so much as ratify a governance effort that's already underway, led not by states or international bodies but by shareholders. When governments fail to meaningfully regulate, and corporations cannot be trusted to do what is beneficial beyond their own bottom line, people in society still have the power to set us on the right path
The argument that shareholders are filling the regulatory vacuum is both empirically interesting and structurally fragile. Shareholder activism depends on institutional investors prioritizing ESG over returns — a position under constant pressure. If fiduciary duty arguments win in court, the entire governance apparatus described here loses its legal standing. The Pope's authority cannot shore up what securities law might undermine.
The government sought to evaluate models up to 90 days prior to release, while AI labs pushed for a much shorter timeline of only 14 days.
大多数人认为科技公司会支持更严格的监管以确保安全，但作者认为AI公司实际上推动的是更短的测试时间，这挑战了科技公司总是寻求更多监管保护的主流观点。
Trump has taken a hands-off approach to regulating AI since retaking office, but members of his administration got spooked and began recommending safety testing after Anthropic flagged cybersecurity risks with its latest model, Mythos.
大多数人认为特朗普政府会继续其宽松的科技监管立场，但作者认为特朗普政府内部出现了分歧，部分官员在安全事件后转向支持AI安全测试，这挑战了人们对特朗普一贯的监管风格的预期。
In practical terms, in the age of AI and robotics, ensuring that the economy favors human dignity means adopting certain criteria for firm action. First, transparency and accountability: when data and algorithms influence credit distribution, personnel selection or access to services and opportunities, it is necessary that decisions be understandable, contestable and subject to oversight, so that individuals are not reduced to mere profiles. Second, inclusion and access: the benefits of innovation must be paired with investments in skills, infrastructure and essential services to ensure that technology does not widen the gap between those who have and those who have not. Finally, measures to ensure equity: taxation, social protection and industrial policies must correct the imbalances created by the concentration of wealth and power. Indeed, these criteria do not constitute a curb on innovation; instead they make it civilized and humane.
Suggests regulation along the lines of algorithmic/data transparency & accountability, investing the profits of innovation in education and essential services, and laws and policies which check the concentration of wealth and power.
The bill would shield frontier AI developers from liability for 'critical harms' caused by their frontier models as long as they did not intentionally or recklessly cause such an incident.
令人惊讶的是：该法案将AI开发者的责任限定在'故意或鲁莽'行为上，这意味着即使AI系统导致大规模死亡或财务灾难，开发者也可能免于承担责任。这种近乎完全的责任豁免条款在产品责任法中极为罕见，反映了AI监管的特殊性。
Sanders and Rep. Alexandria Ocasio-Cortez (D-NY) introduced a bill to ban data center construction "until Congress passes comprehensive AI legislation."
令人惊讶的是：伯尼·桑德斯和亚历山德里娅·奥卡西奥-科尔特斯这两位政治立场截然不同的政治人物竟然联手提出暂停数据中心建设的法案，这表明AI监管问题已经成为跨党派议题，超越了传统政治分歧。
We provide a framework for categorizing the ways in which conflicting incentives might lead LLMs to change the way they interact with users, inspired by literature from linguistics and advertising regulation.
令人惊讶的是：研究人员借鉴语言学和广告监管领域的文献来构建分析框架，这表明AI系统中的利益冲突问题与传统的广告和语言操纵有着深刻的联系，暗示了AI可能正在采用传统广告中的操纵策略。
The government has so far favoured a pro-innovation, sector-led approach, prioritising voluntary principles over hard regulation.
大多数人认为政府会迅速采取立法行动保护创作者权益，但作者指出英国政府实际上倾向于自愿原则而非硬性监管。这一观点挑战了公众对政府会在AI版权问题上采取强硬措施的预期，揭示了政策制定的实际倾向。
The government has so far favoured a pro-innovation, sector-led approach, prioritising voluntary principles over hard regulation.
大多数人认为英国政府在AI监管方面会采取强硬立场保护创作者权益。但作者指出政府实际上倾向于亲创新、行业主导的方法，优先考虑自愿原则而非硬性监管。这一发现与公众对政府保护创作者的期望形成鲜明对比，揭示了政策现实与公众认知之间的差距。
If you told me you were building a next generation nuclear power plant, but there was no way to get accurate readings on whether the reactor core was going to blow up, I’d say you shouldn’t build it. Is A.I. like that power plant? I’m not sure.
This is the weird part of these articles … he has just made a cast-iron argument for regulation and then says "I'm not sure"!!
That first sentence alone is enough for the case. Why? Because he doesn't need to think for sure that AI is like that power plant ... he only needs to think there is a (even small) probability that AI is like that power plant. If he thinks that it could be even a bit like that power plant then we shouldn't build it. And, finally, in saying "I'm not sure" he has already acknowledged that there is some probability that AI is like the power plant (otherwise he would say: AI is definitely safe).
Strictly, this is combining the existence of the risk with the "ruin" aspect of this risk: one nuclear power blowing up is terrible but would not wipe out the whole human race (and all other species). A "bad" AI quite easily could (malevolent by our standards or simply misdirected).
All you need in these arguments is a simple admission of some probability of ruin. And almost everyone seems to agree on that.
Then it is a slam dunk to regulate strongly and immediately.
Who is responsible for the actions of AI? How should liability be determined for their mistakes? Can a legal system designed by humans keep pace with activities produced by an AI capable of outthinking and potentially outmaneuvering them?
Politically, people have been pushing deregulation for decades, but we have regulations for a reason, as these questions illustrate.