Researchers at Anthropic and AI safety company Andon Labs gave an instance of Claude Sonnet 3.7 an office vending machine to run. And hilarity ensued.

A recent experiment from Anthropic shows just how strange things can get when an AI agent tries to do a human job. In "Project Vend," researchers gave their Claude Sonnet 3.7 AI the task of ...

Anthropic let AI Agent Claude run a store in its office for a month to see whether it could turn a profit. The results left lots of room for improvement.