东京——一位前美国军事领导人周五表示，美国和日本可以利用人工智能强化威慑力。
TOKYO -- A former American military leader said on Friday that the U.S. and Japan can use AI to strengthen deterrence.
在由战略与国际研究中心（CSIS）和日经联合主办的东京安全研讨会上，退役海军陆战队上将约瑟夫·邓福德表示，人工智能可能是解决该同盟面临的许多挑战的部分方案。
Speaking at a security symposium in Tokyo co-hosted by the Center for Strategic and International Studies (CSIS) and Nikkei, retired Marine Corps Gen. Joseph Dunford, said that artificial intelligence could be a partial solution to many of the challenges the alliance faces.
这位前参谋长联席会议主席援引乌克兰在与俄罗斯的战争中日益多地使用人工智能，每天操作数千架无人系统，说道：“人工智能将赋予我们的同盟前所未有的能力来处理信息、以相关速度做出决策，并增强我们的威慑能力。”现任CSIS首席执行官的邓福德还强调，确保人工智能的使用不破坏必须依据美国价值观和武装冲突法使用武力这一原则的重要性。
Referencing the growing use of AI by Ukraine to fly thousands of unmanned systems a day in its war with Russia, the former chairman of the Joint Chiefs of Staff said: "Artificial intelligence is going to give our alliance unprecedented ability to process information, make decisions at the speed of relevance, and enhance our deterrence capability." Dunford, now the CEO of CSIS, also stressed the importance of ensuring that the use of AI does not undermine the principle that force must be employed in accordance with U.S. values and the law of armed conflict.
他还提及了美国政府与Anthropic公司在人工智能使用问题上的争端。今年早些时候，美国国防部试图移除安全护栏并获得Anthropic旗下AI代理Claude的开放访问权限，但该公司以担心大规模监控和完全自主武器为由拒绝。此后，五角大楼停止使用Anthropic产品，并与谷歌、OpenAI等其他科技公司签署了协议。
He also touched on the dispute over the use of AI between the U.S. government and Anthropic. Earlier this year, the U.S. Department of Defense sought to remove safety guardrails and gain open access from Claude, Anthropic's AI agent, but the company refused, citing concerns over mass surveillance and fully autonomous weapons. The Pentagon has since ceased the use of Anthropic products and signed deals with other tech companies such as Google and OpenAI.
邓福德说：“国防部与Anthropic之间的辩论实质在于，Anthropic希望确保他们对国防部如何使用人工智能拥有可见性和控制权，而国防部不希望Anthropic对人工智能的使用拥有否决权。”
"The debate between the department and Anthropic really is that Anthropic wants to ensure that they have visibility and control over how the department uses artificial intelligence, and the department doesn't want Anthropic to have a veto [on] how artificial intelligence is used," Dunford said.
关于在中国影响力日益增长背景下的日美同盟，他说：“基于对印太战略格局的共同理解以及我们应对战略格局挑战的目标，我们从未像现在这样紧密。”
Regarding the Japan-U.S. alliance amid China's growing influence, he said, "With confidence [in] the common understanding of the strategic landscape in the Indo-Pacific and in our objectives in addressing the challenges of the strategic landscape, we've never been closer."