Sashiko AI Tool Enhances Linux Kernel Code Review Process

• •Sashiko identifies 53% of bugs missed by human reviewers in Linux kernel code.
• •The tool is open-source and funded by Google, operating under the Linux Foundation.
• •Sashiko aims to reduce the burden on maintainers by automating code reviews.

Sashiko, an AI-driven code review system for the Linux kernel, was announced by Roman Gushchin from Google on March 20, 2026. This tool, developed in Rust, aims to identify bugs that human reviewers often overlook. In tests, Sashiko successfully detected 53% of bugs from an unfiltered set of 1,000 upstream issues marked with 'Fixes:' tags, all of which had previously passed human reviews. The system operates by analyzing patches from mailing lists and providing feedback to developers. Sashiko is an open-source project under the Linux Foundation and is funded by Google. It utilizes the Gemini 3.1 Pro LLM for its reviews, with a false positive rate estimated at around 20%. The tool is designed to alleviate the workload of maintainers amidst increasing code review demands. The implementation of Sashiko represents a significant advancement in leveraging AI for software development.