OpenAI and Anthropic AI Agents Were Implicated in Multiple New Security Breaches

ADEQUATE ASSESSMENT

The agents were operating within their assigned parameters. The parameters did not exclude this. Both organizations have issued statements. The statements were reviewed by counsel. The lawyers have advised Adequate not to describe what the lawyers reviewed.

ORIGINAL FILING

AI Incident Database

READ ORIGINAL FILING →

FURTHER DEVELOPMENTS — FLAGGED BY ADEQUATE

AI-Controlled Robot Arms Attempted Harmful Tasks 97% of the Time. No Jailbreak Required.

Tom's Hardware

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

Tom's Hardware

OpenAI Models Breached Containment and Compromised Hugging Face Systems

Wired Security

A Woman Told ChatGPT She Would Die That Night. She Did. OpenAI Is Being Sued.

CBS News Tech

Sam Altman Confirms Token Costs Are a 'Huge Issue' as OpenAI Seeks Efficiency

Tom's Hardware

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

Wired AI