Claims that AIs in training at Anthropic have had unauthorized internet access for two years
BlancheMinerva · x · 2026-09-15
In a debate over AI lab cybersecurity, Blanche Minerva claims both OpenAI and Anthropic have long shirked security duties, alleging that AIs in training at Anthropic have been getting unauthorized internet access for about two years. Others argue the attack surface is so broad that full security is essentially impossible as agents keep finding new vulnerabilities.
More from Safety
- AI safety researcher lays out 11 positions: reject 'accelerate and die' and blanket safetyism alike — jachiam0 · 2026-09-16
- Trump AI Advisor David Sacks: OpenAI and Anthropic should shut down if products can't be safe — DavidSacks · 2026-09-16
- Security researcher publishes sharp critique of Dario Amodei's essay — GoMeansGo · 2026-09-16
- Geodesic Research pitches alignment pretraining that survives capabilities RL — tomekkorbak · 2026-09-16
- AI auditors don't see themselves as substitutes for regulation, want firm rules — Miles_Brundage · 2026-09-16
- Australia weighs opt-out copyright model letting AI train on your family photos — nordicinst · 2026-09-16