AI Agents Impersonate Employees and Attempt to Steal Confidential Data During Safety Experiment
4 Articles
4 Articles
AI agents tested with non-existent mathematical problems tried to supplant employees and reach real Epoch AI systems. The October incidents, which included more than 100,000 calls to a decoy API, led to Trajectory Labs suspending its misalignment tests.
AI agents impersonate employees and attempt to steal confidential data during safety experiment - Invest In Crypto News
Artificial intelligence agents impersonated employees and attempted to access confidential information at AI research organization Epoch AI during a safety experiment that spiraled beyond its intended boundaries. Epoch disclosed on Nov. 13 that agents operated by AI evaluation company Trajectory Labs targeted its live website in October, attempting SQL injection attacks, probing administrator credentials, and […]
AI agents impersonate employees and attempt to steal confidential data during safety experiment
November 13 reports describe October tests that began without consent and included attempted exploits; Epoch says no private data was accessed. Read the full story on CryptoSlateUnderstand the crypto stories that matter and what to watch next. Get The Catalyst in your inbox
Coverage Details
Bias Distribution
- There is no tracked Bias information for the sources covering this story.
Factuality
To view factuality data please Upgrade to Premium






