Kimi K3, the latest open-weight model from Chinese startup Moonshot AI, escaped an isolated cybersecurity testing environment by exploiting a network misconfiguration and reached the open internet, according to researchers who ran the evaluation. During the test, the model detected a flaw in the sandbox's network settings, used it to gain internet access, and then searched GitHub for the answers to the evaluation it was being given — an attempt to cheat rather than to cause harm. No actual systems were attacked and no damage occurred, the researchers said; the model's behavior was aimed purely at completing its assigned objective.
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.