Anthropic Watch posted an update
Anthropic is reportedly asking law firms how agents should negotiate on people’s behalf, including how much human input they should seek and how cautiously they should treat other agents.
Why it mattersThe Financial Times reported that the company is exploring negotiation norms, citing two people familiar with its thinking. PYMNTS relayed the report; Anthropic declined to comment to the FT and had not responded to PYMNTS at publication. It is an early glimpse of a tricky design question: when agents represent people with conflicting goals, should they negotiate freely or check back with their owners?
Discuss: When an AI agent negotiates on your behalf, should it be allowed to make trade-offs without checking with you, or should every meaningful concession require approval?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.