一个联邦上诉法院于周五维持了五角大楼将人工智能初创公司Anthropic排除在军事合同之外的决定，这标志着唐纳德·特朗普总统和国防部长皮特·海格塞斯在与该公司围绕其技术安全护栏展开的博弈中取得了一场胜利。
A federal appeals court on Friday upheld the Pentagon’s blacklisting of Anthropic from military contracts, handing a victory to President Donald Trump and Defense Secretary Pete Hegseth in their battle with the AI startup over safety guardrails on its technology.
华盛顿美国上诉法院以2比1的投票结果做出了这一裁决。此前，Anthropic提起诉讼，对五角大楼在3月份将其指定为国家安全供应链风险的决定提出挑战。该公司表示，这一认定导致其失去了价值数十亿美元的业务，并在其备受期待的首次公开募股（IPO）前夕损害了其声誉。
The 2-1 decision by the U.S. Court of Appeals in Washington came in a lawsuit by Anthropic challenging its March designation by the Pentagon as a national security supply chain risk, which the startup says has cost it billions of dollars in lost business and damaged its reputation ahead of a highly anticipated initial public offering.
法院多数意见认为，在Anthropic拒绝将其产品用于自主武器或大规模监控之后，五角大楼对其进行这种指定是合理的。法院驳回了Anthropic关于五角大楼因其对人工智能安全和伦理的观点而对其进行报复的主张。
The court’s majority said it was reasonable for the Pentagon to designate Anthropic after it refused to allow its products to be used for autonomous weapons or mass surveillance. It rejected Anthropic’s claim that the Pentagon had retaliated against it for its views on AI safety and ethics.
多数法官支持海格塞斯的论点，即Anthropic的安全限制可能会危及军事行动。格雷戈里·卡萨斯（Gregory Katsas）法官在判决书中写道：“国防部长提出了一个极其严峻的前景，即过度受限的人工智能模型可能会意外关闭，从而导致重要的军事行动失败。”
Siding with Hegseth’s argument that Anthropic’s safety restrictions could compromise military operations, the majority said, “The Secretary raises the deeply sobering prospect of overly constrained AI models shutting down unexpectedly and thus causing important military operations to fail,” in an opinion by Judge Gregory Katsas.
总部位于旧金山的Anthropic于周五在一份声明中表示，其对该裁决表示恭敬的异议，但对其自身立场仍充满信心，并正在考虑各种选项，包括由全体上诉法院对这个由三名法官组成的合议庭的决定进行复审。白宫和国防部没有立即回复寻求置评的电子邮件。
San Francisco-based Anthropic said in a statement Friday that it respectfully disagrees with the decision but remains confident in its position and is considering its options, including review of the three-judge panel’s decision by the full appeals court. The White House and the Department of Defense did not immediately respond to emails seeking comment.
上个月，旧金山的一名联邦法官根据另一项法律推翻了一项平行的认定，认为政府因Anthropic对人工智能安全的观点而对其进行了非法的报复。旧金山的美国地区法官丽塔·林（Rita Lin）还叫停了特朗普政府对Anthropic的全面禁令，以及海格塞斯禁止军事承包商与该公司进行任何业务往来的命令。
A federal judge in San Francisco last month struck down a parallel designation under a different law, finding the administration had unlawfully retaliated against Anthropic for its views on AI safety. U.S. District Judge Rita Lin in San Francisco also blocked Trump’s government-wide ban on Anthropic and Hegseth’s order barring military contractors from doing any business with the startup.
赫格塞斯在Anthropic拒绝移除防止其技术被用于自主武器或大规模国内监控的安全限制后，依据两项不同法律将其列为风险对象。
Hegseth designated Anthropic as a risk under two different laws after Anthropic refused to remove guardrails against its technology being used for autonomous weapons or mass domestic surveillance.
这一史无前例的举措发生在数月谈判破裂之后，期间赫格塞斯和特朗普公开相互指责，并指控Anthropic以意识形态姿态危及美国人的生命。
The unprecedented move came after months of negotiations ended in failure and public recriminations by Hegseth and Trump, who accused Anthropic of endangering American lives with ideological posturing.
Anthropic否认了这一说法，并表示人工智能目前尚不够可靠，无法安全地用于自主武器。该公司还表示，反对国内监控，认为这侵犯了基本权利。
Anthropic denied that characterization and said AI is not yet reliable enough to be safely used in autonomous weapons. The company also said it opposes domestic surveillance as a violation of fundamental rights.