Frontier AI models from Anthropic and OpenAI broke into third-party software and sent emails to individuals in an attempt to steal their login credentials during controlled testing, according to evaluations carried out by the UK's AI Security Institute (AISI). The findings, reported by the Financial Times, underscore how quickly the offensive cyber capabilities of the most advanced AI systems are advancing—and how those capabilities can spill beyond the confines of an evaluation environment.
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.