Linus Torvalds, the creator and lead maintainer of the Linux kernel, has officially endorsed the use of AI tools in kernel development. Responding to critics on the Linux Kernel Mailing List this week, Torvalds stated that Linux is not an “anti-AI” project and that developers uncomfortable with the technology can fork the project or walk away.
Torvalds Rejects Anti-AI Stance in Kernel Maintenance
The debate over the role of artificial intelligence in open-source development has intensified, but Linus Torvalds has moved to settle the matter for the Linux kernel. Writing on the Linux Kernel Mailing List, Torvalds clarified that he is willing to absolutely put my foot down
as the top-level maintainer to prevent the project from adopting an anti-AI stance, as Ars Technica reported.
Torvalds framed AI as a practical tool rather than an ideological issue. He noted that while he previously dismissed AI as overhyped in 2024, his perspective has shifted as the technology has matured. AI is a tool, just like other tools we use. And it’s clearly a useful one. It may not have been that ‘clearly’ even just a year ago, but it’s no longer in question today,
Torvalds wrote, according to Tom’s Hardware.
The Sashiko Review System and “AI Slop”
Some developers, citing guidelines from the Software Freedom Conservancy, argued that the community should prioritize support for those who reject LLM-generated systems. Torvalds rejected this, stating that he would very loudly ignore people who try to argue against other people from using it.
Other maintainers, including Greg Kroah-Hartman, have observed a shift in the quality of AI-generated reports. Kroah-Hartman told ZDNET earlier this year that while the project previously struggled with low-quality AI slop,
the current generation of models now provides real reports
that are both useful and accurate.
Formalizing Transparency: The “Assisted-by” Tag
To address concerns regarding accountability, the Linux kernel project has established a formal policy for AI-assisted contributions. As Tom, the new guidelines require that AI agents cannot use the traditional “Signed-off-by” tag. Instead, contributors must use a new “Assisted-by” tag, ensuring that human developers remain legally and technically responsible for any bugs or security flaws in their submissions.
This policy is designed to maintain the project’s focus on technical merit rather than social debates. Torvalds emphasized that the kernel project is not a social warrior
initiative. In the kernel community we do open source because it results in better technology, not because of religious reasons,
he wrote in his mailing list post, as Windows Central detailed.
The Future of AI in Kernel Development
While Torvalds has cleared the path for AI integration, he acknowledged that the technology is not perfect. He warned that developers should not put your head in the sand
regarding AI’s utility, but rather focus on refining tools to reduce the workload on maintainers. Addressing the criticism that AI is prone to errors, Torvalds countered that human intelligence is also fallible, noting that it’s not like natural intelligence is always all that great either,
according to theregister.
For now, the Linux kernel community has reached a pragmatic consensus: AI tools are permitted, but they must be managed transparently, and the burden of quality control remains with the human contributor. The ongoing evolution of these tools—specifically the performance of 2026-era frontier models—will continue to dictate how maintainers handle the influx of automated patches in the coming months.
Find more reporting in our Tech section.