Anthropic researchers wear down AI ethics with repeated questions
How do you get an AI to answer a question it’s not supposed to? There are many such “jailbreak” techniques, and Anthropic researchers just found a new one, in which a large language model can be convinced to tell you how to build a bomb if you prime it with a few dozen less-harmful questions […] © 2024 TechCrunch. All rights reserved. For personal use only.
![Anthropic researchers wear down AI ethics with repeated questions](https://techcrunch.com/wp-content/uploads/2024/02/GettyImages-1424498694.jpg?resize=1200,800)
How do you get an AI to answer a question it’s not supposed to? There are many such “jailbreak” techniques, and Anthropic researchers just found a new one, in which a large language model can be convinced to tell you how to build a bomb if you prime it with a few dozen less-harmful questions […]
© 2024 TechCrunch. All rights reserved. For personal use only.
What's Your Reaction?
![like](https://todaypic.com/assets/img/reactions/like.png)
![dislike](https://todaypic.com/assets/img/reactions/dislike.png)
![love](https://todaypic.com/assets/img/reactions/love.png)
![funny](https://todaypic.com/assets/img/reactions/funny.png)
![angry](https://todaypic.com/assets/img/reactions/angry.png)
![sad](https://todaypic.com/assets/img/reactions/sad.png)
![wow](https://todaypic.com/assets/img/reactions/wow.png)