Hackers are learning to exploit chatbot ‘personalities’

Groucho Marx glasses on a computer processor.

This is The Stepback, a weekly newsletter breaking down one essential story from the tech world. For more on AI mischief, follow Robert Hart. The Stepback arrives in our subscribers' inboxes at 8AM ET. Opt in for The Stepback here.

How it started

Hacking the first generation of AI chatbots was a laughably simple affair. You didn't need any technical know-how, backdoor access, or even a basic understanding of what a large language model was. You didn't need to code. To get an AI system that had cost billions to build to abandon its safety instructions, sometimes all you had to do was ask.

These attacks, known as jailbreaks, had the quality …

Read the full story at The Verge.



source https://www.theverge.com/column/935545/hackers-ai-chatbots

Comments

Popular posts from this blog

GE made a 27-inch smart display for above your stove that streams Netflix and Spotify

Mophie’s battery pack case for the new iPhones lets you use wired headphones

Vizio returns to CES with its most advanced 4K TV ever and support for Apple’s AirPlay 2