"In my measurement, Sashiko was able to find 53 percent of bugs based on a completely unfiltered set of 1,000 recent upstream issues based on 'Fixes:' tags (using Gemini 3.1 Pro). Some might say that 53 percent is not that impressive, but 100 percent of these issues were missed by human reviewers."
"The quality of reviews is high... the rate of false positives is harder to measure, but based on limited manual reviews it's well within 20 percent range, and the majority of it is a gray zone."
"We've been using it internally at Google for some time, and it helped to discover a large number of real issues."
Sashiko is a Rust-based AI tool designed for code review in the Linux kernel, announced by Roman Gushchin. It identifies bugs by analyzing patches from mailing lists, achieving a 53 percent detection rate of issues previously missed by human reviewers. The tool aims to alleviate the workload of maintainers amidst increasing code reviews. While it has a false positive rate within 20 percent, its effectiveness has been validated through internal use at Google. Sashiko operates under the Linux Foundation and is supported by Google for its operational costs.
Read at Theregister
Unable to calculate read time
Collection
[
|
...
]