Google has announced that its artificial intelligence-based tool named Big Sleep has discovered 20 security weaknesses in the popular open-source software. The company shared this development as part of its widespread initiative to use AI in cyber security research. Heather Edkins, vice president of the security of security, revealed the update through a post on X, earlier known as Twitter.
Google AI Equipment Problems in Popular devices
The Big Sleep developed in collaboration between Google Deepmind and Project Zero analyzed a variety of open-source tools and detected flaws in systems such as FFMPEG and Imantemagick. These devices are widely used in multimedia processing, including to handle audio, video and image files. Although Google has not released specific technical details of the weakened weakened, it confirmed that the AI system settles and reproduced the flaws independently. A human safety analyst later reviewed and verified the findings before giving information to the company.
Also read: Xiaomi unveiled new AI Voice Model to promote auto, home tech
How big sleep is operated
The AI system works by imitating malicious users’ behavior and scanning software code and network services for weak points. Big sleep not only examines the system for weaknesses, but also adopts its ways over time and learns new ways to highlight complex issues. 20 flaws have been discovered so far in Google’s internal platforms and many open-sources projects.
Also read: CEO Tim Cook says that Apple is ready to open his wallet to catch in AI
According to Google, the purpose of deploying AI in this capacity is not to replace human safety researchers, but to help them. The company highlighted that its AI agent could execute thousands of test cases much faster than humans, which would allow cyber security teams to focus on strategic decisions, while AI manages regular testing.
Also read: Samsung Galaxy S25 Review: Flagship features in a handful of package
Big Sleep is not the only AI tool involved in such work. Other AI systems like Runsybil and XBow have also shown a promise in the bug-skill sector. XBOW recently led a leaderboard on the Hacaron Bug Bounty Platform in the United States.
Also read: Oppo F29 Pro Review: Polish artist who matters it matters
While AI devices have started reopening how weaknesses are discovered, concerns remain. Some developers take care of an increase in wrong bug reports generated by AI, often called “AI Slope”. These reports sometimes have errors or imaginary flaws. However, some experts from the region, including Runsybil’s co-founder Vlad INESCU, have supported Google’s Big Sleep as a serious project supported by correct expertise and infrastructure.
Mobile finder: Google Pixel 9 Pro XL latest price, glasses and all details
