Anthropic trains AI agents to negotiate like lawyers

Anthropic is researching how to train AI agents to negotiate contracts, salaries, and prices by mimicking professional legal standards. The company is reportedly consulting with law firms to analyze how attorneys advocate for client interests and seek equitable outcomes, aiming to integrate these professional norms into AI behavior.

As autonomous agents are expected to handle increasing numbers of commercial and personal disputes, Anthropic is studying interaction norms between different AI models. Key areas of focus include determining when an agent should request human intervention and establishing an appropriate level of skepticism when interacting with other AI systems to prevent dysfunctional or overly distrustful behavior.

Experimental data indicates a strong correlation between an AI model's capabilities and its negotiation success. In studies involving the exchange of goods via AI proxies, higher-performing models consistently secured better results for their users, exhibiting social dynamics similar to human negotiators.