From Skeptic to Power User: Linus Torvalds Actively Writes Kernel Patches with Claude Code

The creator of Linux has moved beyond defending artificial intelligence from the sidelines and is now actively using it to write serious code. In a move that underscores a rapid evolution in his perspective, Linus Torvalds recently submitted a hefty patch series to the Linux kernel mailing list, openly crediting an LLM coding assistant for the heavy lifting.

On October 12, 2026, Torvalds posted an eight-patch series under a Git branch explicitly named noref-filemap-read-with-llm. The update aims to optimize how Linux handles page cache reads across multiple threads, addressing a performance bottleneck that dates back to 2024. The technical improvement itself was quickly overshadowed by Torvalds' candid admission. "The code, the test programs, the review behind the list above and the first draft of this text were done with an LLM coding assistant (Claude Code) under my direction," he wrote to the mailing list.

This hands-on integration of AI connects directly to the decisive stance Torvalds took a few months prior. In July 2026, amid rising friction over automated tooling in the kernel code review process, Torvalds firmly put his foot down. As previously covered by AI Night Watch, he declared unequivocally that "Linux is not one of those anti-AI projects," telling critics that AI serves as a practical and useful utility. At the time, he pushed back against individuals arguing for a blanket rejection of these technologies, maintaining that Linux development is driven by technical merit and unburdened by ideological fears of new tools.

Torvalds’ trajectory over the last couple of years reveals a pragmatic, merit-based evolution. In late 2024, he famously dismissed the AI industry as "90 percent marketing and 10 percent reality," viewing it mostly as a potential future aid for catching bugs. As the technology matured, his approach adapted. His July 2026 defense established the philosophical groundwork for embracing AI as an undeniable utility that maintainers should harness. Now, his October patch series represents the practical execution of that philosophy. By explicitly naming his branch after an LLM and crediting Claude Code for both the implementation and the documentation, Torvalds is leading by example.

Torvalds has seamlessly transitioned from a cautious skeptic to a defender of AI tooling, and finally to an active AI co-developer. For an open-source community that closely monitors its founder's every move, this evolution sends a clear message: artificial intelligence has officially arrived as a daily, functional instrument for writing the Linux kernel.