'Please do not vibe f--- up this software': Broken backups spark AI coding row in rsync project

AI and ml

Users probe backup failures find Claude-assisted commits. Veteran engineer retorts: "I did not just vibe-code 'convert test suite to python'."

Incremental backups started failing for some rsync users after a recent update, and what they found in the project's commit history quickly turned a routine bug hunt into yet another fight over AI-generated code.

The controversy centers on rsync 3.4.3, a security-focused release published earlier this year to fix multiple vulnerabilities. Shortly after the upgrade, some users reported that incremental backup workflows were no longer behaving as expected, with one user saying their backup system failed on anything other than a full backup.

Rsync creator Andrew Tridgell has pushed back against the criticism in a Medium post titled "Rsync and Outrage," arguing that many commenters have drawn conclusions without understanding how the AI tools were actually used.

Rsync is not a weekend side project maintained by three people in a Discord server. First released in the 1990s, it remains one of the most widely used file synchronization and backup utilities in the Unix and Linux world. Countless backup products, scripts, NAS appliances, and IT departments depend on it quietly doing its job without surprises.

That makes any suggestion of AI-assisted development in the project far more contentious than it might be elsewhere.

The backup issue might have remained a fairly ordinary bug report had users not started poking around in rsync's recent commit history. They found that since rsync 3.4.1, dozens of commits have been attributed to "tridge and claude," referring to rsync creator Andrew Tridgell and Anthropic's AI assistant Claude.

The discovery prompted a strongly worded GitHub post titled "Please Do Not Vibe Fuck Up This Software," a reference to the increasingly common practice of handing coding tasks to AI models and trusting the results.

From there, the discussion spread to Reddit and Hacker News, where the conversation shifted from a backup bug to a broader debate about AI-generated code finding its way into critical open source infrastructure.

Veteran developer Tridgell acknowledged that rsync 3.4.3 introduced regressions affecting some backup workflows, describing them as "valid (but unusual) use cases" that were not covered by the project's existing test suite. "I apologize if your use case of rsync was hit by these regressions," he wrote.

But Tridgell pushed back on suggestions that he had simply handed development over to Claude and hoped for the best.

According to Tridgell, the most visible AI-assisted work involved rewriting rsync's aging shell-script test suite in Python as part of a broader effort to improve security testing and harden the codebase. He said he designed the framework himself, used Claude alongside OpenAI's Codex and Google's Gemini for what he described as "grunt work," and manually reviewed the resulting code.

"I did not just vibe-code 'convert test suite to python,'" he wrote. "I'm a software engineer with 40 years experience."

Tridgell also argued that maintainers are increasingly dealing with a flood of security reports, many of them AI-generated, which has dramatically increased the workload required to keep widely used open source software secure.

"The world of software engineering has changed dramatically in the last few months," he wrote. "The world of IT security and maintaining software in the face of the flood of reports has completely and utterly changed just in the last few weeks."

Far from backing away from AI-assisted development, Tridgell suggested he intends to continue using the tools as rsync heads toward a larger 3.5 release focused on security improvements. He also took a swipe at users threatening to jump ship to OpenBSD's openrsync project, noting that rsync's new test suite currently reports dozens of failures when run against the alternative implementation.

Whether that reassurance satisfies critics is still unclear. But if nothing else, the whole thing demonstrates that AI-assisted development and backup software make for a combustible combination. One involves trusting a machine – the other exists because people don't. ®