Sign up to save your podcasts
Or
Share A New Trick Uses AI to Jailbreak AI Models—Including GPT-4
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
A New Trick Uses AI to Jailbreak AI Models—Including GPT-4
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 about your ad choices: dovetail.prx.org/ad-choices
View all episodes
By WIRED
4
5858 ratings
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 about your ad choices: dovetail.prx.org/ad-choices
More shows like Security, Spoken
View all
Security Now (Audio)
1,983 Listeners
Uncanny Valley | WIRED
448 Listeners
SANS Internet Stormcenter Daily Cyber Security Podcast (Stormcast)
636 Listeners
CyberWire Daily
1,009 Listeners
What's New
100 Listeners
Science, Spoken
56 Listeners
Business, Spoken
50 Listeners
Smashing Security
312 Listeners
Click Here
415 Listeners
Malicious Life
925 Listeners
Cybersecurity Today
166 Listeners
Hacking Humans
314 Listeners
TechCrunch Daily Crunch
42 Listeners
WIRED Politics Lab
234 Listeners
Cyber Security Headlines
127 Listeners