Hackers are learning to exploit chatbot ‘personalities’

May 24, 2026 Robert Hart

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.

Previous Article
Apple’s latest MacBook Air is $200 off in both sizes for Memorial Day
Apple’s latest MacBook Air is $200 off in both sizes for Memorial Day

If you’re looking for a laptop that’ll still perform well several years from now, Apple’s latest MacBook Ai...

Next Article
Why Nuro thinks being a robotaxi ‘second mover’ gives it an advantage
Why Nuro thinks being a robotaxi ‘second mover’ gives it an advantage

Waymo is the undisputed leader in the robotaxi space, operating a fleet of over 3,000 driverless cars in at...