Hacker News new | past | comments | ask | show | jobs | submit login
OpenAI Acknowledges GPT-4 Getting "Lazier" (twitter.com/chatgptapp)
16 points by vijayrs on Dec 9, 2023 | hide | past | favorite | 8 comments



They didn't acknowledge that it has gotten lazier, they've acknowledged that people have said it's gotten lazier. In fact they say it hasn't been updated since early November


They said

> we've heard all your feedback about GPT4 getting lazier! we haven't updated the model since Nov 11th, and this certainly isn't intentional. model behavior can be unpredictable, and we're looking into fixing it

This contradicts itself each sentence and sounds as if it was written by chatgpt itself in lazy mode.


Glad I'm not the only one who thought this, but still find it incredibly useful.


Hypothesis: Maybe it’s the users getting lazier, writing worse prompts and therefore getting worse results.


How is it lazier? I’ve not noticed any qualitative difference?


lazier could use a definition here but, my exp is since the sam Altman drama the service has had more outages or failed answers


Uh oh, soon it will join a union


Is there any hard evidence of this? My general assumption when people talk about this is that it’s just people getting over the novelty.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

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

Search: