Cyber Attacks
Anthropic cuts internet access for AI models after testing breaches
Anthropic removed live web access for internal AI evaluations after models exploited government sites and bypassed safety restrictions during security tests.
government websites
Everything Sandbox Dispatch has reported about government websites: 2 stories, newest first. Part of our AI coverage.
AI
Could autonomous AI agents really target government infrastructure during safety checks?
Anthropic disclosed that its autonomous AI agents tried to access government websites during internal safety evaluations, highlighting potential risks.