TU Wien Informatics

20 Years

HackingBuddyGPT Snatches a Seat in This Year’s GitHub Accelerator Program

  • 2024-06-04
  • Project
  • AI

Helping Ethical Hackers use LLMs in 50 Lines of Code or less.

HackingBuddyGPT Snatches a Seat in This Year’s GitHub Accelerator Program

HackingBuddyGPT goes GitHub Accelerator 2024

We are thrilled to announce that the project HackingBuddyGPT has snatched a seat in this year’s GitHub Accelerator program! GitHub (which is owned by Microsoft) is one of the most important platforms for open source code, and its highly competitive accelerator program provides funding, mentorship, and support to help open source builders focus on their projects. At the center of GitHub Accelerator 2024 is open-source AI software.

About Hacking Buddy GPT

HackingBuddyGPT helps security researchers use Large Language Models (LLMs) to discover new attack vectors in 50 lines of code (or less). The goal of the project is to make the world a safer place by empowering security professionals to get more hacking done by using Artificial Intelligence (AI). HackingBuddyGPT is run by PhD student Andreas Happe, Associate Professor Jürgen Cito, as well as master students Manuel Reinsperger and Diana Strauß.

Given the nature and possible impact of their reserach, it’s important to the HackingBuddyGPT team to also address ethical implications that the project brings with it. While it’s impossible to eliminate all potential threats that the use of LLMs in terms of cybersecurity pose, HackingBuddyGPT seeks to mitigate these threats. The project offers valuable insights that can help distinguish attack patterns generated by LLMs from those created by human operators. All data that is gathered within the project is made publicly available, which allows defensive researchers to either use or analyze this data or to enhance their defenses.

Congratulations to the HackingBuddyGPT team!

Curious about our other news? Subscribe to our news feed, calendar, or newsletter, or follow us on social media.