ITs Mr Sweary Linus Torvalds says AI tools are helping Linux development, while stuffing maintainers’ inboxes with duplicate bug-report slop.
According to Techstrong Linux creator Torvalds told the Linux Foundation’s Open Source Summit North America that AI has changed kernel development fast.
“In the last six months, we’ve seen a lot more commits… the last two releases, it’s been about 20 per cent more commits than we had in the previous releases over many years.”
That sounds like the sort of productivity bump AI boosters love to print on conference lanyards. The less glossy bit is what happens when maintainers must sort the machine-made mess.
“The real change that happened in the last six months was that the AI tools actually got good enough for a lot of people… we’re seeing a definite uptick in just development on pretty much all fronts….” Torvalds said.
Torvalds is not doing the usual tech priest routine of declaring AI either salvation or Satan. He seems to think it is useful, interesting and deeply irritating.
“I have a love-hate relationship with AI. I actually really like it from a technical angle, I love the tools, I find it very useful and interesting, but it is definitely causing pain points,” Torvalds said.
One useful bit is bug discovery. If AI spots a real fault in source code, the short-term nuisance can still produce a better kernel.
“When AI finds a bug in any source code… long term is you found a bug, we fixed it, that the end result is better for it. I think finding bugs is great, because the real problem is all the bugs you didn’t find…” he said.
The trouble starts when AI-generated reports arrive as drive-by dumps. A small maintainer team then has to separate useful findings from automated confetti.
Torvalds said this is especially rough for small teams and solo maintainers. They get a report, ask for more information and the submitter has already wandered off.
“it’s a bug report, and when you ask for more information, the person has done a drive-by and doesn’t even answer your questions anymore,” Torvalds said.
The Linux security mailing list appears to have taken the brunt of the AI bug rush. Not every bug is a security issue, but that has not stopped people from lobbing everything at the security list.
“People think that when they find a bug with AI, the first reaction sometimes seems to be let’s send it to the security list, because this may have security implications,” Torvalds said.
That left a small group of maintainers wading through duplicate entries. AI did not replace the work so much as move it onto the poor sods who still answer emails.
The Linux project has responded with tools and updated documentation. Part of that paperwork now addresses “an uptick in bug and security reports from discoveries made in full or in part with AI.”
One tool, Sashiko, reviews patches submitted on the mailing list. It is not perfect, but Torvalds said it can spot issues and ask the obvious questions before humans lose the will to live.
