Project Swap: What happens when agents trade for us? \ Anthropic
anthropic.com · 6,154 words · saved by 2 readers
To see what works and what breaks when agents are sent into a market, we made a miniature market of Claudes.
Summary To see what works and what breaks when agents are sent into a market, we made a miniature market of Claudes—a more controlled sequel to Project Deal, our first experiment with agents interacting in a marketplace on people's behalf. Anthropic employees across six offices brought in a book they wanted to give away. Each participant had a short chat with Claude about what they like to read, and sent a Claude-powered agent onto an open trading floor to pitch, haggle, and strike deals with other people’s agents. The goal was for everyone to take home a summer read they would enjoy.…
saved by
related reading
- Project Deal: our Claude-run marketplace experiment | Anthropic \ Anthropicanthropic.com
- Project Vend: Can Claude run a small shop? (And why does that matter?) \ Anthropicanthropic.com
- Patterns and problems in multiagent systemsanthropic.com
- Tracing the thoughts of a large language model \ Anthropicanthropic.com
- Automated Weak-to-Strong Researcheralignment.anthropic.com
- How we built our multi-agent research system \ Anthropicanthropic.com
- Demystifying evals for AI agents \ Anthropicanthropic.com
- Teaching Claude Whyalignment.anthropic.com
- How Claude Performs on Robotics Tasks \ Anthropicanthropic.com
- AGI Tradesdanielgross.com
- Coasean Bargaining at Scaleblog.cosmos-institute.org
- Building Effective AI Agents \ Anthropicanthropic.com