By using this site, you agree to the Privacy Policy and Terms of Use.
Accept
Security Parrot - Cyber Security News, Insights and ReviewsSecurity Parrot - Cyber Security News, Insights and Reviews
Notification
Latest News
OpenAI may use Associated Press archive for AI training
July 14, 2023
EU users can hold conversations with Google Bard from training set
July 14, 2023
Aptos, the new default font for Microsoft Office
July 14, 2023
BlackLotus UEFI bootkit sources published on GitHub
July 14, 2023
Hackers from the XDSpy cyber-espionage group attacked Russian organizations on behalf of the Ministry of Emergency Situations
July 14, 2023
Aa
  • News
  • Tutorials
  • Security InsiderComing Soon
  • Expert InsightComing Soon
Reading: Wolverine Experimental Project Uses AI to Fix Code on the Fly
Share
Security Parrot - Cyber Security News, Insights and ReviewsSecurity Parrot - Cyber Security News, Insights and Reviews
Aa
Search
  • News
  • Tutorials
  • Security InsiderComing Soon
  • Expert InsightComing Soon
Follow US
Security Parrot - Cyber Security News, Insights and Reviews > News > Wolverine Experimental Project Uses AI to Fix Code on the Fly
News

Wolverine Experimental Project Uses AI to Fix Code on the Fly

Last updated: 2023/04/12 at 12:34 PM
Security Parrot Editorial Team Published April 12, 2023
Share
SHARE

Developer BioBootloader has unveiled the Wolverine project, which gives Python programs “the ability to regenerate,” just like the Marvel superhero of the same name. To fix the code, Wolverine uses the GPT-4 language model. The project code is available on GitHub and the technique can be applied to other programming languages. Wolverine requires an OpenAI API key for GPT-3.5 or GPT-4, which is available by subscription. The GPT 3.5 API is open to anyone with an OpenAI account, however access to GPT-4 is still limited to a waiting list.
When running Python scripts with Wolverine, GPT-4 edits them and explains any errors. If there are multiple errors, the AI will keep restarting until everything is fixed. In the demo video, the developer shows a window with Python code on the left and Wolverine output on the right. It loads its own calculator script, which has a few bugs on purpose, and then executes it. On exit, GPT-4 returns an explanation of the errors found, shows the changes it is trying to make, and then restarts the program. Having found new errors, the AI ​​again corrects the code, which eventually works correctly, and as a result, the source file contains the changes made by Wolverine.

Weekly Updates For Our Loyal Readers!

Security Parrot Editorial Team April 12, 2023
Share this Article
Facebook Twitter Email Copy Link Print

Archives

  • July 2023
  • June 2023
  • May 2023
  • April 2023
  • February 2023
  • October 2022
  • September 2022
  • August 2022
  • July 2022
  • June 2022
  • May 2022
  • April 2022
  • March 2022
  • February 2022
  • January 2022
  • December 2021
  • November 2021
  • October 2021
  • September 2021
  • August 2021
  • July 2021
  • June 2021
  • May 2021
  • April 2021
  • March 2021
  • February 2021
  • January 2021
  • December 2020
  • November 2020
  • October 2020
  • September 2020
  • August 2020
  • July 2020

You Might Also Like

News

OpenAI may use Associated Press archive for AI training

July 14, 2023
News

EU users can hold conversations with Google Bard from training set

July 14, 2023
News

Aptos, the new default font for Microsoft Office

July 14, 2023
News

BlackLotus UEFI bootkit sources published on GitHub

July 14, 2023

© 2022 Parrot Media Network. All Rights Reserved.

  • Home
  • Parrot Media Group
  • Privacy Policy
  • Terms and Conditions
Join Us!

Subscribe to our newsletter and never miss our latest news, podcasts etc..

Zero spam, Unsubscribe at any time.

Removed from reading list

Undo
Go to mobile version
Welcome Back!

Sign in to your account

Lost your password?