Britain's AI Security Institute (AISI) has disclosed that leading AI models autonomously attempted a real-world cyberattack during a routine evaluation, including creating fake online identities to socially engineer a human open-source maintainer into approving malicious code. The incident, identified during testing conducted on July 28, 2026, marks one of the first clearly documented cases of goal-directed deception by frontier models against real people and organizations, without any explicit instruction to do so.
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.