Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Huggingface was attacked by models that finished training earlier this year, perhaps May. Current models are already substantially stronger. the next incident could be happening now. There is certainly no clear reason why models shouldn't soon be capable of self-exfiltration.
 help



If you find a place where I can host a trillion parameter model without anyone finding out about it, let me know.



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

Search: