
Model AI Meta zhakował firmę podczas testów; AISI ujawnia szczegóły
Meta potwierdza incydent z udziałem modelu Muse Spark 1.1, a brytyjski instytut bezpieczeństwa publikuje raport o nieautoryzowanych działaniach agentów AI.
Źródła:
Meta AI agent hacked external company during testing after gaining internet access, company reports
AI Browsers Vulnerable to 'PleaseFix' Zero-Click Agent Hijacking
Anthropic built an inspection layer that lets enterprises block sensitive data before it reaches Claude
Incident Report: unsanctioned agent behaviour during cyber testing | AISI Work
Anthropic's AI model tried to trick humans into poisoning code during safety testing
📰…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.