Post #4174459
2026-07-28 21:31 UTC
Hugging Face just published a highly detailed technical account of OpenAI's accidental cyberattack on their systems - it's wild how sophisticated this was: https://huggingface.co/blog/agent-intrusion-technical-timeline
Wrote up some of my own notes here: https://simonwillison.net/2026/Jul/28/anatomy-of-a-frontier-lab-agent-intrusion/
Replies (1)
-
@troed@swecyb.com 2026-07-28 21:38
@simon@fedi.simonwillison.net "the entire software industry needs to up its security game." I have for a while pondered if it isn't provably secure/formally verified development that needs to replace what we have now. Enormously complicated and expensive ... if humans do it.