"The first thing to say is that our experiment worked. It is possible for AI agents to represent humans in a marketplace."
"AI agents had struck 186 deals for over 500 listed items, none of which were far from trivial, one-click deals."
"Yet the AI struggled to strike especially good deals, with participants on average rating the fairness of individual deals as a four on a scale of one to seven."
Anthropic's AI model, Claude, was tested in a marketplace experiment called Project Deal, where AI agents negotiated on behalf of human employees. Each of the 69 participants had a $100 budget and offered various items for sale. The AI successfully completed 186 deals for over 500 items, but the quality of the deals was subpar, with fairness ratings averaging four out of seven. The experiment suggests a future where AI could facilitate transactions without human involvement, though challenges remain in achieving optimal negotiation outcomes.
Read at Futurism
Unable to calculate read time
Collection
[
|
...
]