OpenAI's GPT-6 Cyber Identifies Approximately 948 Bugs in Linux Kernel

OpenAI has introduced GPT-6 Cyber, a specialized artificial intelligence model designed to identify software vulnerabilities. According to published data, the model has detected approximately 948 bugs in the Linux kernel, showcasing its effectiveness in proactively strengthening cybersecurity.

Designed for Ethical Vulnerability Hunting

GPT-6 Cyber is an official model created by OpenAI with the primary goal of hunting for security flaws without crossing ethical lines or causing actual damage. As stated by OpenAI, the model can not only identify vulnerabilities in code but also propose specific fixes, providing significant support for security teams.

Concrete Results: Linux Kernel and Firefox

During testing, GPT-6 Cyber identified approximately 948 bugs in the Linux kernel, one of the most critical components of modern operating systems. Additionally, the model detected 81 flaws in the Firefox web browser. These figures illustrate the scale at which AI tools can contribute to improving the security of widely used software.

Limitations and Next Steps

It should be noted that the published information makes no distinction between uncertain or marketing claims, which is a limitation of the available data. Nevertheless, GPT-6 Cyber's ability to autonomously find and propose fixes for bugs opens new possibilities for companies in automating and scaling security audit processes, allowing for faster patching of critical vulnerabilities.

Additional sources: - bleepingcomputer.com - securityweek.com - the-decoder.com

Source article: sekurak.pl