Linus Torvalds uses Claude Code on experimental Linux kernel changes
Linus Torvalds used Claude Code to develop eight experimental Linux kernel changes, but none has entered the official tree.
Linux creator Linus Torvalds has begun using Anthropic’s Claude Code to work on experimental changes intended to address a Linux kernel performance issue, although none of the changes have entered the official kernel yet. 1
What Torvalds used Claude Code for
According to Computerworld, Torvalds used Claude Code to develop eight code changes and accompanying test programmes under his supervision. The work is documented as an experiment rather than a finished kernel contribution. 1
The report says the code, tests, review and first draft of the accompanying documentation were produced with help from the AI coding assistant. Torvalds still supervised the process, and the changes remain outside the official Linux kernel. 1
Why it matters
Torvalds has historically been sceptical of AI-generated code, but recent comments suggest his view has softened. That makes the experiment notable: it is not evidence that Linux development has switched wholesale to AI, but it is a clear example of the kernel’s creator testing where the tools can help. 12
Our opinion
The interesting part is not that an AI tool touched kernel code; it is that Torvalds kept the work experimental, reviewed and separate from the official tree. That is the sensible standard for high-risk infrastructure: use the tool where it saves time, but make human review and upstream acceptance the real test. This is a meaningful experiment, not proof that AI can independently maintain Linux.
The bottom line
The eight changes have not been merged into the official Linux kernel. For now, this is a closely watched experiment in AI-assisted systems programming rather than a change to how Linux is maintained. 12