The Decoder · 2026/10/18 17:11:29
Anthropic 咨询律所制定 AI 谈判规范
Anthropic 正与律师事务所合作，为未来代表用户进行价格谈判、处理投诉或修改合同条款的 AI 智能体制定公平协商规范。此举旨在应对此前实验中发现的“更强模型能达成更优交易”可能加剧社会不平等的风险，并试图解决法律等人文领域缺乏可验证数据以优化模型判断力的难题。
报道全文原始报道全文
Anthropic is figuring out how AI agents could negotiate on behalf of users. According to the Financial Times, the company is consulting with law firms to develop fair negotiation norms for AI agents that would haggle over prices, resolve complaints, or improve contract offers. Anthropic has already shown in experiments that stronger models negotiate better deals, which could widen societal inequality.
This isn't new territory. Anthropic is already hiring Conceptual Reasoning Fellows, specifically lawyers. The fellows will spend four months testing Claude on judgment-heavy questions in law, philosophy, cognitive science, economics, and political science for 167,000.
The push traces back to a data problem. In math and code, improving AI models is relatively easy because verifiable data is plentiful or easy to generate. In the humanities, that data doesn't exist, partly because right and wrong answers are rarely clear-cut. Anthropic needs experts who can build test cases and evaluations to make Claude stronger in these areas. Justin Curl, a member of technical staff researching law, institutions, and AI at Anthropic, argues that in law it's "notoriously hard to define quality."