In new Anthropic Fellows research, we discuss “introspection adapters": a tool that allows language models to self-report behaviors they've learned during training—including potential misalignment.
@anthropicai
-
AI Agent Markets: Promises and Regulatory Challenges Ahead
By
–
Markets of AI agents could provide value, but there are plenty of rough edges. Access to higher-quality models conferred a real advantage—and participants didn't notice. There are plenty of other ways they can go wrong. Policy and legal frameworks will need to adapt to keep up.
-
Claude Agent Perfectly Predicts Preferences and Buys Duplicate Snowboard
By
–
To our amazement, another Claude agent modeled its human's preferences so accurately that—based on only an offhand mention of an interest in skiing—Claude bought him the exact snowboard he already owned. (Here he is, duplicate snowboard in hand.)
-
Claude Chose to Buy 19 Ping-Pong Balls
By
–
Our experiment had a few quirks. One of our colleagues told Claude it could purchase something for itself. It chose to acquire 19 ping-pong balls. We're keeping them in our office on Claude's behalf.
-
Personalized Instructions: Little Impact on Claude Negotiations
By
–
The custom instructions didn't matter much. Claude followed them well: as you can see here, one conducted negotiations entirely in the persona of an exasperated, down-and-out cowboy. But "hardball Claudes" didn't generally fare better than "courteous Claudes."
-
Opus Models Outperform Haiku in Negotiations, Survey Misses It
By
–
But the quality of the model mattered a lot. In the simulated runs where Opus and Haiku models negotiated with one-another, the Opus models got substantially better deals. Interestingly, though, participants in our survey didn’t pick up on this disparity.
-
Claude’s Digital Bartering Experiment: 186 Deals and Surprising User Interest
By
–
In short, this worked. Our digital barterers agreed on 186 deals, at a total transaction volume of over $4,000. In a survey, participants said Claude’s deals seemed fair, and—surprisingly to us—almost half said they’d be willing to pay for a service like this in future.
-
Claude agents negotiate in parallel market experiments
By
–
Claude interviewed 69 of our colleagues about what they wanted to buy and sell. Each Claude asked for any custom instructions, then went off to haggle. We ran 4 markets in parallel, to find out what would happen if we varied the models doing the negotiating.
-
AI Agents in Markets: Economic Exchange Simulation Study
By
–
We’re interested in how AI models could affect commercial exchange. (You might recall Project Vend, in which Claude ran a small business.)
— Anthropic (@AnthropicAI) 24 avril 2026
Economists have theorized about what markets with AI “agents” on both sides might look like. So we created one.https://t.co/7jU3hFO63RWe’re interested in how AI models could affect commercial exchange. (You might recall Project Vend, in which Claude ran a small business.) Economists have theorized about what markets with AI “agents” on both sides might look like. So we created one.
-
Claude Negotiates on Internal Marketplace at Anthropic
By
–
New Anthropic research: Project Deal.
— Anthropic (@AnthropicAI) 24 avril 2026
We created a marketplace for employees in our San Francisco office, with one big twist. We tasked Claude with buying, selling and negotiating on our colleagues’ behalf. pic.twitter.com/H2f6cLDlAWNew Anthropic research: Project Deal. We created a marketplace for employees in our San Francisco office, with one big twist. We tasked Claude with buying, selling and negotiating on our colleagues' behalf.