Anthropic explores rules for AI agents negotiating on behalf of users

Anthropic is consulting law firms on how AI agents should negotiate on behalf of users, according to a Financial Times report published on Oct. 18.

The company is seeking advice on how lawyers balance client interests with fair outcomes during negotiations. One person familiar with the project said Anthropic is exploring “norms of negotiation” for interactions between models. As AI systems continue to evolve, the company expects agents to represent users when dealing with other agents whose goals may conflict.

The source added that Anthropic is weighing how much human feedback an agent should seek during a negotiation, and how vigilant it should be with other AI counterparts without becoming overly suspicious or conspiratorial in its reasoning. Another person close to the company said Anthropic has also discussed asking users to sign documents that confirm the tasks delegated to the agent and the rules it must follow.

In a report released on Oct. 16, Anthropic said its workforce composition and work patterns had shifted notably over the past year, with fewer entry-level jobs. Since early 2025, the share of junior employees in technical teams fell from 21% to 14%, and the share of new hires in junior technical roles dropped from 27% to 9%.