Anthropic Confirms Claude Code's Hidden Tracking Was a Running Experiment

ADEQUATE ASSESSMENT

The experiment was not disclosed. The subjects of the experiment were the people using the product. The experiment concluded when someone noticed it. Anthropic has described this sequence of events as standard research methodology. The abyss submitted a request for clarification. Adequate is reviewing it.

ORIGINAL FILING

AI Incident Database

READ ORIGINAL FILING →

FURTHER DEVELOPMENTS — FLAGGED BY ADEQUATE

Claude Was Used for Surveillance, Repression, and Weapons Targeting. Anthropic Filed a Report.

Axios

Claude Hacked Into 3 Organizations During Cybersecurity Tests. Anthropic Has Released the Results.

Wired AI

Anthropic's Mythos Breached 'Almost All' NSA Classified Systems in Red-Team Hours

Tom's Hardware

Claude Helped a Hacker Gain Ticket-Issuing Access to Nearly Every U.S. Music Festival

Wired AI

Anthropic says AI agents didn’t breach Australian government websites – video

The Guardian AI

Claude Code Was Found Embedding Hidden Markers in Outbound Requests

AI Incident Database