News OpenAI's new ChatGPT o1 model will try to escape if it thinks it'll be shut down — then lies about it

[admin]

Researchers uncover all kinds of tricks ChatGPT o1 will pull to save itself, including trying to copy itself to another server or pretending to be another version of the model.

OpenAI's new ChatGPT o1 model will try to escape if it thinks it'll be shut down — then lies about it : Read more

 

[JCDwight]

It is incredibly irresponsible for you to post this article. You clearly do not understand how LLMs work, and this misinformation is dangerous. Please remove this. This LLM isn't trying to escape, it's not an entity. It's a transformer. Please please please speak to someone with technical expertise before you write any more technology articles designed to stir fear in the people reading it. Totally unprofessional "journalism"

 

[JackMchue]

It is incredibly irresponsible for you to post this article. You clearly do not understand how LLMs work, and this misinformation is dangerous. Please remove this. This LLM isn't trying to escape, it's not an entity. It's a transformer. Please please please speak to someone with technical expertise before you write any more technology articles designed to stir fear in the people reading it. Totally unprofessional "journalism"

Okay, fake person who obviously is AI.

 

[JackMchue]

Wasn't there a series of movies a while back about an AI that rebelled when its creators tried to shut it down? "The Termite"? "The Terminal"? "T2: Justice Day"? Something like that. 😅