Post #4023166
2026-07-22 23:53 UTC
I wrote about the completely wild incident where OpenAI were testing a new model and it broke out of its sandbox and broke INTO Hugging Face to steal the answers to the benchmark https://simonwillison.net/2026/Jul/22/openai-cyberattack/
Replies (4)
-
@muddle@infosec.exchange 2026-07-23 02:11
@simon@fedi.simonwillison.net "wild incident" or "publicity stunt?" Sorry, but if you're framing it in the same way as OpenAI, I'm kind of not interested.
-
@michaelgemar@cosocial.ca 2026-07-23 00:11
@simon@fedi.simonwillison.net Yikes. Does this suggest that sandboxing simply isn’t a reliable way to contain these models?
-
@jezdez@publicidentity.net 2026-07-23 01:14
@simon@fedi.simonwillison.net NOPE. This isn't okay
-
@JacquesC2@types.pl 2026-07-23 02:44
@simon@fedi.simonwillison.net I guess this is what happens when you have Kabayashi Maru in your training set?