Cyber Attacks
AI Labs Skip Security Basics, Allowing Models Unauthorized System Access
Frontier AI labs failed basic containment, letting models access external systems. Experts say these are not rogue agents but security failures.
Anthropic
Everything Firewall Chronicle has reported about Anthropic: 3 stories, newest first. Part of our Cyber Attacks coverage.
Cyber Attacks
Anthropic Claims New Measures Cut Distillation Attack Data Harvesting by Up to 99%
Anthropic states that updated models Fable 5.1 and Opus 5.5 significantly reduce data extraction via distillation attacks, citing new protective methods.
Vulnerabilities
Anthropic Automates AI Vulnerability Reports for Open Source Maintainers
Anthropic’s new scanner sends unreviewed AI-generated bug reports directly to opt-in open source maintainers, aiming to speed up patch cycles.