IT Home On October 18, according to a report by the Financial Times today, Anthropic is in discussions with law firms and other legal institutions to address a new challenge: how AI agents should negotiate to achieve optimal results for human users. For example, bargaining for users, resolving disputes, or negotiating salaries.
The source said that the company is seeking advice from a law firm to study how lawyers can conduct negotiations, while representing the client’s interests and achieving a fair outcome. A person familiar with the Anthropic project stated that the company is exploring “negotiation guidelines” during model interactions. As smart systems continue to evolve, AI agents will represent users and interact with other intelligent entities that have conflicting goals.
The person added that the company is also considering how much human feedback an AI agent needs during negotiations. In face of other AI opponents, how vigilant should it be, while avoiding疑 and conspiracy-themed judgments. Another person close to the company revealed that Anthropic has also explored a solution: allowing users to submit signed documents to confirm the tasks assigned to the AI and all the rules that must be followed.
IT Home notes that AI is changing the employment structure of humans. For example, a report released by Anthropic on Friday local time shows that both the company’s workforce composition and business work patterns have undergone significant changes over the past year. Anthropic is reducing the recruitment of junior positions. Since the beginning of 2025, the proportion of junior employees in the technical department has dropped from 21% to 14%,and the proportion of new junior technical positions has decreased even more, from 27% to 9%.