Anthropic consults law firms on AI agent negotiation norms
Anthropic is consulting law firms on how AI agents should negotiate on users' behalf, aiming for norms that pursue good deals while preserving fair outcomes, according to Financial Times reporting relayed by 36Kr, IT之家 and The Decoder. People familiar with the project described debates over how much human feedback agents should seek, how wary they should be of other AI agents, and a possible signed mandate from users spelling out the agent's instructions and rules. The Decoder also reported that Anthropic is hiring lawyers as Conceptual Reasoning Fellows to test Claude on judgment-heavy questions and said the company's experiments showed stronger models negotiate better deals.
The reported work addresses how agents should represent users' interests while dealing with other AI agents, a rules question the FT says tech companies face as machines act on humans' behalf.
Anthropic聘请法律人士,助力AI智能体完成交易谈判-36氪
登录
搜索