Linus Torvalds Declares War on Pointless AI-Generated Links in Linux Kernel Submissions
Share this article
Linux creator Linus Torvalds unleashed a blistering critique this week targeting a pervasive annoyance in kernel development: automated, context-free links in code submissions. The outburst—posted on the Linux Kernel Mailing List (LKML)—followed a trivial patch submission that included a "Link:" argument promising explanation but instead looped back to redundant information.
"Stop this garbage already. Stop adding pointless Link arguments that waste people's time. Add the link if it has ADDITIONAL information. Dammit, I really hate those pointless links... they only make my job harder," Torvalds fumed, emphasizing that his core role involves "trying to make sense of pull requests."
Why Pointless Links Infuriate Maintainers
Torvalds isn't alone in his frustration. Maintainers routinely sift through hundreds of patches, and AI-generated links—often auto-added by platforms like GitHub or GitLab—frequently point to:
- Duplicate summaries instead of bug reports
- Generic issue trackers with no technical context
- Automated changelogs lacking critical discussion history
This forces maintainers to hunt for actual rationale, wasting precious time. As one Redditor noted: "That original patch looked like AI-written summary, and the link... was the same summary."
The AI Automation Dilemma
The surge in "link spam" coincides with broader adoption of AI coding tools. While designed to streamline workflows, these systems often prioritize compliance over utility:
1. Scripts/Bots: Auto-embed references without verifying value
2. Template-Driven Workflows: Junior developers follow rigid protocols
3. Superficial Compliance: Links added to "tick boxes" rather than aid review
Torvalds explicitly called out this trend: "Can we please stop this automated idiocy?"
A Watershed Moment for Open Source
The episode underscores a critical challenge: as AI infiltrates development, meaningful communication risks being sacrificed for speed. Torvalds' outburst serves as a stark reminder that maintainers—especially in complex projects like the Linux kernel (now at 40M+ lines of code)—require clear, human-driven context to evaluate changes safely.
For developers, the takeaway is unambiguous: links must illuminate, not obfuscate. Submissions should prioritize:
- Links to specific bug reports or design discussions
- Concise summaries explaining why a change matters
- Elimination of automated references lacking unique insights
As AI reshapes coding, Torvalds’ stance reinforces that tools should augment—not replace—thoughtful collaboration. The survival of open source’s review culture depends on it.
Source: ZDNet