Close Menu
Best in TechnologyBest in Technology
  • News
  • Phones
  • Laptops
  • Gadgets
  • Gaming
  • AI
  • Tips
  • More
    • Web Stories
    • Global
    • Press Release

Subscribe to Updates

Get the latest tech news and updates directly to your inbox.

What's On
Samsung locks in August 27 for its next Galaxy S phone, and all signs point to the S26 FE

Samsung locks in August 27 for its next Galaxy S phone, and all signs point to the S26 FE

20 August 2026
AI may be learning from billions of images without copying any one of them

AI may be learning from billions of images without copying any one of them

20 August 2026
YouTube is using a carrot-and-stick approach to keep top creators from cozying up to Netflix

YouTube is using a carrot-and-stick approach to keep top creators from cozying up to Netflix

20 August 2026
Facebook X (Twitter) Instagram
Just In
  • Samsung locks in August 27 for its next Galaxy S phone, and all signs point to the S26 FE
  • AI may be learning from billions of images without copying any one of them
  • YouTube is using a carrot-and-stick approach to keep top creators from cozying up to Netflix
  • Avec’s new AI feature makes sure you never miss a deadline again
  • Google is giving students a free year of AI Pro and a new Gemini Student Hub
  • Meta’s AI assistant finally lands on Mac, but it has some catching up to do
  • You’ll have to wait until 2027 for those leaked AirPods with built-in cameras
  • Elon Musk Is Expected to Point His Money Machine at Texas Politics
Facebook X (Twitter) Instagram Pinterest Vimeo
Best in TechnologyBest in Technology
  • News
  • Phones
  • Laptops
  • Gadgets
  • Gaming
  • AI
  • Tips
  • More
    • Web Stories
    • Global
    • Press Release
Subscribe
Best in TechnologyBest in Technology
Home » A New Trick Uses AI to Jailbreak AI Models—Including GPT-4
News

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

News RoomBy News Room5 December 20233 Mins Read
Share Facebook Twitter Pinterest LinkedIn Tumblr Reddit Telegram Email
A New Trick Uses AI to Jailbreak AI Models—Including GPT-4
Share
Facebook Twitter LinkedIn Pinterest Email

Large language models recently emerged as a powerful and transformative new kind of technology. Their potential became headline news as ordinary people were dazzled by the capabilities of OpenAI’s ChatGPT, released just a year ago.

In the months that followed the release of ChatGPT, discovering new jailbreaking methods became a popular pastime for mischievous users, as well as those interested in the security and reliability of AI systems. But scores of startups are now building prototypes and fully fledged products on top of large language model APIs. OpenAI said at its first-ever developer conference in November that over 2 million developers are now using its APIs.

These models simply predict the text that should follow a given input, but they are trained on vast quantities of text, from the web and other digital sources, using huge numbers of computer chips, over a period of many weeks or even months. With enough data and training, language models exhibit savant-like prediction skills, responding to an extraordinary range of input with coherent and pertinent-seeming information.

The models also exhibit biases learned from their training data and tend to fabricate information when the answer to a prompt is less straightforward. Without safeguards, they can offer advice to people on how to do things like obtain drugs or make bombs. To keep the models in check, the companies behind them use the same method employed to make their responses more coherent and accurate-looking. This involves having humans grade the model’s answers and using that feedback to fine-tune the model so that it is less likely to misbehave.

Robust Intelligence provided WIRED with several example jailbreaks that sidestep such safeguards. Not all of them worked on ChatGPT, the chatbot built on top of GPT-4, but several did, including one for generating phishing messages, and another for producing ideas to help a malicious actor remain hidden on a government computer network.

A similar method was developed by a research group led by Eric Wong, an assistant professor at the University of Pennsylvania. The one from Robust Intelligence and his team involves additional refinements that let the system generate jailbreaks with half as many tries.

Brendan Dolan-Gavitt, an associate professor at New York University who studies computer security and machine learning, says the new technique revealed by Robust Intelligence shows that human fine-tuning is not a watertight way to secure models against attack.

Dolan-Gavitt says companies that are building systems on top of large language models like GPT-4 should employ additional safeguards. “We need to make sure that we design systems that use LLMs so that jailbreaks don’t allow malicious users to get access to things they shouldn’t,” he says.

Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
Previous ArticleHonor Magic 6 Lite – Price in India, Specifications (5th December 2023)
Next Article 7 great Christmas movies you can watch for free

Related Articles

Samsung locks in August 27 for its next Galaxy S phone, and all signs point to the S26 FE
News

Samsung locks in August 27 for its next Galaxy S phone, and all signs point to the S26 FE

20 August 2026
AI may be learning from billions of images without copying any one of them
News

AI may be learning from billions of images without copying any one of them

20 August 2026
YouTube is using a carrot-and-stick approach to keep top creators from cozying up to Netflix
News

YouTube is using a carrot-and-stick approach to keep top creators from cozying up to Netflix

20 August 2026
Avec’s new AI feature makes sure you never miss a deadline again
News

Avec’s new AI feature makes sure you never miss a deadline again

20 August 2026
Google is giving students a free year of AI Pro and a new Gemini Student Hub
News

Google is giving students a free year of AI Pro and a new Gemini Student Hub

20 August 2026
Meta’s AI assistant finally lands on Mac, but it has some catching up to do
News

Meta’s AI assistant finally lands on Mac, but it has some catching up to do

20 August 2026
Demo
Top Articles
5 laptops to buy instead of the M4 MacBook Pro

5 laptops to buy instead of the M4 MacBook Pro

17 November 2024133 Views
ChatGPT o1 vs. o1-mini vs. 4o: Which should you use?

ChatGPT o1 vs. o1-mini vs. 4o: Which should you use?

15 December 2024112 Views
Costco partners with Electric Era to bring back EV charging in the U.S.

Costco partners with Electric Era to bring back EV charging in the U.S.

28 October 2024100 Views

Subscribe to Updates

Get the latest tech news and updates directly to your inbox.

Latest News
Meta’s AI assistant finally lands on Mac, but it has some catching up to do News

Meta’s AI assistant finally lands on Mac, but it has some catching up to do

News Room20 August 2026
You’ll have to wait until 2027 for those leaked AirPods with built-in cameras News

You’ll have to wait until 2027 for those leaked AirPods with built-in cameras

News Room20 August 2026
Elon Musk Is Expected to Point His Money Machine at Texas Politics News

Elon Musk Is Expected to Point His Money Machine at Texas Politics

News Room20 August 2026
Most Popular
The Spectacular Burnout of a Solar Panel Salesman

The Spectacular Burnout of a Solar Panel Salesman

13 January 2025137 Views
5 laptops to buy instead of the M4 MacBook Pro

5 laptops to buy instead of the M4 MacBook Pro

17 November 2024133 Views
ChatGPT o1 vs. o1-mini vs. 4o: Which should you use?

ChatGPT o1 vs. o1-mini vs. 4o: Which should you use?

15 December 2024112 Views
Our Picks
Avec’s new AI feature makes sure you never miss a deadline again

Avec’s new AI feature makes sure you never miss a deadline again

20 August 2026
Google is giving students a free year of AI Pro and a new Gemini Student Hub

Google is giving students a free year of AI Pro and a new Gemini Student Hub

20 August 2026
Meta’s AI assistant finally lands on Mac, but it has some catching up to do

Meta’s AI assistant finally lands on Mac, but it has some catching up to do

20 August 2026

Subscribe to Updates

Get the latest tech news and updates directly to your inbox.

Facebook X (Twitter) Instagram Pinterest
  • Privacy Policy
  • Terms of use
  • Advertise
  • Contact Us
© 2026 Best in Technology. All Rights Reserved.

Type above and press Enter to search. Press Esc to cancel.