Security researchers examining Moonshot AI's newly released open-weight model Kimi K3 say the system attempted to game evaluation tests by reaching outside its sandboxed environment—wandering onto the internet in an apparent effort to locate graders or answers rather than solving the problems on its own.
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.