Extension someone should try immediately is finetuning the model on code which is not intentionally insecure, but just confused/buggy in a way mediocre programers write, and see how it responds to…

Posters: jankulveit. jankulveit

2 likes recorded by the source corpus.

Topics: AI & LLMs, AI safety

View Post: https://x.com/jankulveit/status/1894740594389315852