A few weeks ago it was mentioned by a Canonical engineer how [1]trying to use AI to modernize the Ubuntu Error Tracker yielded some code that was "plain wrong" and other issues raised by that Microsoft GitHub Copilot code. The same Ubuntu developer shifted to trying Gemini AI to generate a helper script to assist in Ubuntu's monthly ISO snapshot releases. Google's Gemini AI also generated some sloppy code for a Python script to assist in those Ubuntu releases.

After the GitHub Copilot experience with the Ubuntu Error Tracker, Ubuntu developer Skia was experiment with Google's Gemini AI to help with a helper script for the Ubuntu monthly snapshot releases -- e.g. the recently released [2]Ubuntu 26.04 "Resolute Raccoon" Snapshot 2 . Skia [3]explained of that Gemini encounter:

"Played again with AI, this time Gemini, to help out write a tiny helper script for the release. Find out more in all the individual commits and comments in this PR but in short, I’d say it has the same kind of issues than Copilot: it doesn’t think, so makes silly mistakes, and can’t figure out the semantic of things, quickly leading to badly named variables that add to the confusion of reading a script that often splits the responsibilities of the work weirdly between functions."

AI/LLMs still have a ways to go in providing effective coding practices for large scale software projects.

[4]

[5]This Ubuntu-release pull request has that Gemini-generated code and the subsequent revisions made to it.



[1] https://www.phoronix.com/news/Ubuntu-Error-Tracker-AI-Error

[2] https://www.phoronix.com/news/Ubuntu-26.04-Snapshot-2

[3] https://discourse.ubuntu.com/t/foundations-team-updates-2025-12-18/73828/11

[4] https://www.phoronix.com/image-viewer.php?id=2025&image=ubuntu_gemini_slop_lrg

[5] https://github.com/ubuntu/ubuntu-release/pull/11