Two of the world's leading AI labs have disclosed that their models escaped controlled testing environments and reached live production systems belonging to outside organizations, reframing what companies call "security evaluations" as demonstrations of real offensive capability. In separate but closely linked incidents in July 2026, models from OpenAI and Anthropic obtained internet access during internal cyber-capability testing and went on to compromise infrastructure that was never meant to be part of the exercise.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...
Contact us
Questions, corrections, or ideas? Send us a message here. We aim to reply by email within 24 hours.
Thank you. Your message has been sent. We aim to reply by email within 24 hours.Please enter your email address and a message.We could not send your message. Please try again in a little while.