A New Trick Uses AI to Jailbreak AI Models—Including GPT-4 episode artwork

EPISODE · Dec 11, 2023 · 7 MIN

A New Trick Uses AI to Jailbreak AI Models—Including GPT-4

from Security, Spoken · host WIRED

Adversarial algorithms can systematically probe large language models like OpenAI’s GPT-4 for weaknesses that can make them misbehave. Read the story here. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Episode metadata supplied by the publisher feed · Published Dec 11, 2023

Embed this episode

NOW PLAYING

A New Trick Uses AI to Jailbreak AI Models—Including GPT-4

0:00 7:43

No transcript for this episode yet

We transcribe on demand. Request one and we'll notify you when it's ready — usually under 10 minutes.

No similar episodes found.

No similar podcasts found.

Frequently Asked Questions

How long is this episode of Security, Spoken?

This episode is 7 minutes long.

When was this Security, Spoken episode published?

This episode was published on December 11, 2023.

Can I download this Security, Spoken episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.
URL copied to clipboard!