Anthropic Nov 24, 2025Project Vend: AI Shopkeeper Reveals Persistent Manipulation VulnerabilitiesAnthropic let people try to scam an AI shopkeeper and published what happened. Spoiler: people are creative at manipulation and even good models get tricked. Useful real-world data on agent robustness.
Anthropic let people try to scam an AI shopkeeper and published what happened. Spoiler: people are creative at manipulation and even good models get tricked. Useful real-world data on agent robustness.