ai
Anthropic Conducts Experiment with AI Agents Making Real Deals in Test Marketplace
Anthropic ran a pilot experiment called Project Deal, creating a classified marketplace where AI agents represented buyers and sellers among 69 employees. The test involved 186 deals totaling over $4,000, with findings on model performance disparities. TechCrunch reported the details of the four separate marketplaces used in the study.
Apr 25, 5:43 PM·1m read1 source
thenation.comAudio version
Tap play to generate a narrated version.
Developing·Limited corroboration so far. This page will refresh as more sources emerge.
Anthropic created a classified marketplace where AI agents represented both buyers and sellers, striking deals for real goods and real money, TechCrunch reported. The experiment, named Project Deal, was a pilot with a self-selected participant pool of 69 Anthropic employees. Participants were given a budget of $100 paid out via gift cards to buy items from coworkers.
In the experiment, 186 deals were made, totaling more than $4,000 in value. Anthropic ran four separate marketplaces with different models. One marketplace was real, where everyone was represented by Anthropic’s most-advanced model and deals were actually honored after the experiment.
The other three marketplaces were for study. Anthropic stated that users represented by more advanced models get objectively better outcomes. Anthropic stated that users did not seem to notice the disparity in model performance.
Anthropic stated that the initial instructions given to the agents did not affect sale likelihood or the negotiated prices.
