Linus Torvalds Draws Line: AI Is A Tool For Linux Kernel, Dismisses Non-Technical Objections

Linus Torvalds, the venerable creator of Linux, has unequivocally affirmed the role of Artificial Intelligence as a valuable tool within kernel development, drawing a firm line against those with non-technical objections. This stance comes amidst the increasing effectiveness of AI in critical development workflows, exemplified by tools like Sashiko, an agentic AI code review system. Sashiko, utilizing models such as Gemini 3.1 Pro, has demonstrated significant capability in identifying bugs—reportedly finding 53.6% of previously missed defects in upstream commits, often exceeding human-driven review in detection. Lead Linux kernel maintainer Greg Kroah-Hartman further corroborated this paradigm shift, noting an “overnight” improvement in the quality of AI-generated bug reports from “junk to legit” in late 2023/early 2024, a development he attributes to the wider adoption and refinement of advanced LLMs like Opus 4.5. This accelerated feedback loop is enhancing developer productivity and mitigating security risks, as AI can flag issues faster than human maintainers.

Torvalds, acting as the project’s benevolent dictator for life (BFDL), stated unequivocally that “Linux is not one of those anti-AI projects” and advised those with principal-based, non-technical issues to “fork it or just walk away.” He vehemently dismissed arguments rooted in ethical or social concerns, likening them to the Free Software Foundation’s historically contentious positions, and reiterated that the kernel community makes decisions based solely on “technical merit, not fear of new tools.” This perspective underscores a pragmatic embrace of AI, even in the face of initial resistance, mirroring the adoption curve of technologies like TypeScript. While acknowledging that top-tier human developers might still surpass AI in their specialized niches, the tools significantly raise the “floor” of code quality and efficiency for a broader range of contributors, enabling faster development cycles and proactive identification of vulnerabilities in an increasingly complex software landscape.