Hacker News (curated)new | past | comments | ask | show | jobs| show hidden

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 …


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

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



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact | github