Hacker News new | ask | show | jobs
by hug 44 days ago
Actually, OpenAI made a similar claim about one of their GPT models a while ago…

Funnily enough that was while Dario Amodei was their research director.

1 comments

If you're referring to gpt-2 in 2019, that primarily about concerns with it being used by spammers and fake content generators. In retrospect, that was a totally valid concern.
They had a reddit with GPT2 back and forth I have to say I got suckered into a conversation before I figured it out -- it was definitely the OG Moltbook of non sequiturs