内容来源：https://www.theverge.com/news/885773/anthropic-department-of-defense-dod-pentagon-refusal-terms-hegseth-dario-amodei

内容总结：

五角大楼最后通牒时限将至，人工智能公司Anthropic拒绝军方无限制访问要求

在美国国防部（五角大楼）发出最后通牒、要求其提供人工智能系统无限制访问权限后，知名AI研究公司Anthropic于截止期限前不到24小时正式予以拒绝，坚持其两大原则底线：不参与针对美国民众的大规模监控，以及不开发或提供用于完全自主致命性武器（即“杀手机器人”）的技术。

此次公开对峙是双方近期一系列声明、社交媒体交锋及幕后谈判的高潮。据悉，国防部长皮特·赫格塞思意图重新谈判所有AI实验室与军方的现有合同。在OpenAI与xAI等公司据报已同意新条款的情况下，Anthropic成为唯一公开拒绝者，导致其首席执行官达里奥·阿莫代本周被召至白宫与赫格塞思会面。会上，部长向阿莫代发出最后通牒，要求其在周五营业时间结束前让步，否则将面临后果。

阿莫代在周四晚间的声明中强调，公司“深信利用AI保卫美国及其他民主国家、击败专制对手具有至关重要的意义”，并已主动向国防部及情报界提供模型。但他同时指出，在“少数特定情况下，我们认为AI可能损害而非捍卫民主价值观”，明确点出大规模国内监控与全自主武器系统。他解释，当前前沿AI系统的可靠性尚不足以支撑全自主武器，但并未完全排除未来合作的可能性。

作为回应，五角大楼已被曝要求主要国防承包商评估对其Claude模型的依赖程度，这可能是将其列为“供应链风险”的第一步——此类定性通常仅用于国家安全威胁。此外，军方据称正考虑动用《国防生产法》强制Anthropic配合。

阿莫代在声明中明确表示，五角大楼的“威胁不会改变我们的立场：基于良知，我们无法同意其要求”。他同时承诺，若国防部决定终止合作，公司将确保平稳过渡至其他服务商，避免影响军事规划与行动，并将在其提出的广泛条款下继续提供模型访问。

此次事件凸显了人工智能伦理与国家安全需求之间的尖锐矛盾，亦反映出科技企业在参与国防项目时面临的重大原则抉择。随着最后期限迫近，双方僵局如何收场备受关注。

中文翻译：

五角大楼发出最后通牒的截止时间前不足24小时，Anthropic公司正式拒绝美国国防部对其人工智能系统提出无限制访问的要求。

Anthropic拒绝五角大楼新条款，在致命性自主武器与大规模监控问题上坚守立场

首席执行官达里奥·阿莫代伊的声明发布于五角大楼最后通牒截止时间前不足24小时。这场公开声明、社交媒体帖文与幕后谈判的激烈交锋已进入白热化阶段，其根源在于国防部长皮特·赫格塞思意图重新协商所有人工智能实验室与军方的现有合约。但截至目前，Anthropic始终拒绝在两条现有红线上退让：禁止对美国民众进行大规模监控，禁止研发致命性自主武器（或无需人类监管即可击杀目标的武器）。据报道，OpenAI与xAI已同意接受新条款，而Anthropic的拒绝导致其首席执行官达里奥·阿莫代伊本周被传唤至白宫与赫格塞思会面。据称部长在会面中向这位CEO发出最后通牒，要求其在周五营业时间结束前让步，否则将面临后果。

阿莫代伊在周四晚间的声明中写道："我深信运用人工智能保卫美国及其他民主国家、击败专制对手具有关乎存亡的重要意义。因此Anthropic始终积极推动将我们的模型部署至战争部及情报机构。"他补充说明，公司"从未对特定军事行动提出异议，也未曾试图以临时方式限制技术使用"，但"在少数特定情况下，我们认为人工智能可能侵蚀而非捍卫民主价值观"——随后他特别提及国内大规模监控与全自主武器系统。（阿莫代伊提到"部分自主武器对民主防御至关重要"，全自主武器未来可能"被证明对国防具有关键意义"，但"当前尖端人工智能系统的可靠性尚不足以支撑全自主武器运作"。他未排除Anthropic未来默许军方使用全自主武器的可能性，但强调目前技术尚未成熟。）

据报道，五角大楼已要求主要国防承包商评估其对Anthropic公司Claude系统的依赖程度，这被视为将该公司列为"供应链风险"的第一步——五角大楼近期曾公开作出此类威胁（该分类通常仅用于国家安全威胁）。另有消息称国防部正考虑援引《国防生产法》强制Anthropic就范。

阿莫代伊在声明中强调，五角大楼的"威胁不会改变我们的立场：我们无法昧着良心同意其要求"。他同时表示："若国防部决定终止与Anthropic的合作，我们将全力协助平稳过渡至其他服务商，确保现行军事规划、行动及其他关键任务不受干扰。我们的模型将继续以我们提出的宽泛条款提供所需时长的服务。"

热门资讯

• 汉堡王将启用AI监测员工是否使用"请"和"谢谢"
• 实测体验：三星S26 Ultra全新隐私显示屏令人惊艳
• 谷歌三星联手推出苹果Siri未能实现的AI功能
• 可替代实体钥匙的智能锁标准终落地
• 海盗船宣布3月25日起终止Drop平台销售

英文来源：

Less than 24 hours before the deadline in an ultimatum issued by the Pentagon, Anthropic has refused the Department of Defense’s demands for unrestricted access to its AI.
Anthropic refuses Pentagon’s new terms, standing firm on lethal autonomous weapons and mass surveillance
CEO Dario Amodei’s statement came less than 24 hours before the deadline in the Pentagon’s ultimatum.
CEO Dario Amodei’s statement came less than 24 hours before the deadline in the Pentagon’s ultimatum.
It’s the culmination of a dramatic exchange of public statements, social media posts, and behind-the-scenes negotiations, coming down to Defense Secretary Pete Hegseth’s desire to renegotiate all AI labs’ current contracts with the military. But Anthropic, so far, has refused to back down from its two current red lines: no mass surveillance of Americans, and no lethal autonomous weapons (or weapons with license to kill targets with no human oversight whatsoever). OpenAI and xAI had reportedly already agreed to the new terms, while Anthropic’s refusal had led to CEO Dario Amodei being summoned to the White House this week for a meeting with Hegseth himself, in which the Secretary reportedly issued an ultimatum to the CEO to back down by the end of business day on Friday or else.
In a statement late Thursday, Amodei wrote, “I believe deeply in the existential importance of using AI to defend the United States and other democracies, and to defeat our autocratic adversaries. Anthropic has therefore worked proactively to deploy our models to the Department of War and the intelligence community.”
He added that the company has “never raised objections to particular military operations nor attempted to limit use of our technology in an ad hoc manner” but that in a “narrow set of cases, we believe AI can undermine, rather than defend, democratic values” — going on to specifically mention mass domestic surveillance and fully autonomous weapons. (Amodei mentioned that “partial autonomous weapons … are vital to the defense of democracy” and that fully autonomous weapons may eventually “prove critical for our national defense,” but that “today, frontier AI systems are simply not reliable enough to power fully autonomous weapons.” He did not rule out Anthropic acquiescing to the military’s use of fully autonomous weapons in the future but mentioned that they were not ready now.)
The Pentagon had already reportedly asked major defense contractors to assess their dependence on Anthropic’s Claude, which could be seen as the first step to designating the company a “supply chain risk” – a public threat that the Pentagon had made recently (and a classification usually reserved for threats to national security). The Pentagon was also reportedly considering invoking the Defense Production Act to make Anthropic comply.
Amodei wrote in his statement that the Pentagon’s “threats do not change our position: we cannot in good conscience accede to their request.” He also wrote that “should the Department choose to offboard Anthropic, we will work to enable a smooth transition to another provider, avoiding any disruption to ongoing military planning, operations, or other critical missions. Our models will be available on the expansive terms we have proposed for as long as required.”
Most Popular

• Burger King will use AI to check if employees say ‘please’ and ‘thank you’
• Hands on: I’m super impressed with the Galaxy S26 Ultra’s new Privacy Display
• Google and Samsung just launched the AI features Apple couldn’t with Siri
• The smart lock standard that could replace your keys is finally here
• Corsair is halting Drop sales after March 25th