this post was submitted on 14 Jan 2024
45 points (100.0% liked)

ChatGPT

9002 readers
1 users here now

Unofficial ChatGPT community to discuss anything ChatGPT

founded 2 years ago
MODERATORS
marcar
45
Once an AI model exhibits 'deceptive behavior' it can be hard to correct, researchers at OpenAI competitor Anthropic found (www.businessinsider.com)
submitted 1 year ago by ooli to c/chatgpt
3 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[–] gibmiser 10 points 1 year ago (1 children)

Learned behaviors are hard to unlearn...

[–] MsPenguinette 7 points 1 year ago (1 children)

Once it's learnt this, it'll just get better at lying when you try to punish/correct lies

[–] Mozingo 4 points 1 year ago

Which is exactly what the article says happens