AI programming · 2026-11-10

AI Now Writes More Linux Kernel Fixes Than Bug Reports

Linus Torvalds says that in the latest Linux kernel test release, AI-generated fixes outnumbered automatically generated bug reports, which may signal a shift in AI's role in open-source development.

✨ What happened

In the Linux kernel test release 7.4-rc2, the number of fixes created with artificial intelligence exceeded the number of automatically generated bug reports. Linus Torvalds, founder and chief architect of Linux, reported this on the Linux Kernel Mailing List.

The Linux kernel is the base of many operating systems, so changes to it are noticed worldwide. Torvalds said this may be a sign that the situation is starting to change.

📌 Why it matters for open source

Recently, open-source projects have been struggling with floods of poor bug reports generated by AI bots. Some communities even had to shut down bug bounty programs. Developers spend time sorting through junk.

The Linux kernel handles this better thanks to a higher bar for contributions and a moderation team that filters out weak submissions. Torvalds stressed that AI-written code is acceptable if it passes standard quality checks.

🔹 What changed

Most changes in release 7.4-rc2 are small fixes of one or two lines. Most of them touch graphics, sound and network drivers, as well as the Berkeley Packet Filter (BPF) subsystem. In BPF, the number of AI-generated patches officially exceeded the number of incoming bug reports.

Earlier, Torvalds published an experimental series of eight patches created with the help of the Claude Code AI assistant. It aims to fix a long-standing bottleneck in kernel file caching that significantly slows down parallel data reading.

Sources: 🔗 DOU/dev.ua🔗 ITC.ua