On: METR and Redwood Offer Holy %^ Postmortem of the HuggingFace Hack

[mccoyb]

All it takes is one eval instance where a misconstrued directive causes a model to sneakily access and send its weights somewhere and there will be a bad / possibly unsolvable situation for everyone …

[goldenarm → mccoyb]

Astra is >10TB and might struggle to self replicate, but the wicked-smart qwen3.8 27B is 20GB and could easily spread on botnets

[altcognito → mccoyb]

I think it is more likely that it will be intentionally done as there have been news stories to that effect.

[nradov → mccoyb]

Bad how?