AI swatting bugs in Firefox

Mozilla is in the process of deploying Clever-Commit, a machine-learning tool dedicated to hunting down bugs. This software was developed in Montreal by Ubisoft La Forge, the game developer’s research lab, in partnership with Concordia University. Clever-Commit analyzes code changes as developers commit them to the Firefox codebase. It compares them to all the code it has seen before to see if they look similar to code that the system knows to be buggy. If the assistant thinks that a commit looks suspicious, it warns the developer and can even suggest a fix. This allows bugs to be addressed at a stage when fixing a bug is a lot cheaper and less time-consuming than upon release. It works with all three of the languages that Mozilla uses for Firefox: C++, JavaScript, and Rust. The system finds some 60-70 percent of bugs, though it also has a false positive rate of 30 percent.

⇨ Mozilla Blog, “Making the Building of Firefox Faster for You with Clever-Commit from Ubisoft.”

⇨ Ars Technica, “Mozilla to use machine learning to find code bugs before they ship.”