Advanced

Does terrible code drive you mad? Wait until you see what it does to OpenAI's GPT-4o

Posted by Scott_Ruecker on · Post a comment
The Register; By Thomas Claburn
Mail this story
Print this story

Model was fine-tuned to write vulnerable software – then suggested enslaving humanity. Computer scientists have found that fine-tuning notionally safe large language models to do one thing badly can negatively impact the AI’s output across a range of topics.

Full Story

  Nav
» Read more about: Groups: AI; Story Type: News Story

» Get new stories by e-mail: nightly or as they are published

« Return to the newswire homepage

  Related stories
» More like this

Comments

No comments yet. Be the first.

Post a comment

A free account is needed to post. Write your comment here first: your text is kept in this browser while you sign up or log in, and is waiting for you when you come back.

Your message:
|||||
10
Markdown: **bold**, *italic*, [text](https://link), - lists, > quotes, $x^2$ for math. Formatting help

Sign up and post (it is free)  or  Log in and post