Rémi Verschelde, a key maintainer of the open-source game engine Godot and co-founder of the company W4 Games, stated that the increasing flow of low-quality pull requests generated by large language models is exhausting and demoralizing for Godot developers reviewing changes. According to Rémi, the Godot project prides itself on being open to newcomers and gives any user the chance to influence the development of the engine. The maintainers of Godot spend a lot of time helping to bring pull requests up to standard, but with the current abundance of questionable changes, Rémi does not know how much longer the team can endure.
Currently, there are over 4600 open pull requests accumulated in the Godot repository on GitHub. Adriaan de Jongh, the director of the game development studio Hidden Folks, characterized the situation with AI-generated junk pull requests for Godot as a complete mess and a massive waste of time for reviewers, who are forced to spend time sorting through changes that often make no sense, are excessively verbose in their descriptions, and are submitted by developers who do not understand their own patches.
With the advent of AI tools, maintainers are additionally forced to determine whether the code was written by a person or generated automatically, as well as to assess whether the author understands the code they submitted. They also have to figure out whether the authors tested their code or if the presented test results were merely fabricated by AI. Moreover, Rémi does not understand how to handle situations where, when asked about using AI, the authors of the patches respond, 'I only used it for the description because I write poorly in English,' which does not allow for clear classification of the contribution.
It is noted that maintainers often have to double-check pull requests from new contributors multiple times a day, doubting their quality. The only solution for maintaining the previous approach to development, according to Rémi, is to increase project funding so that they can pay additional maintainers to handle the AI-induced workload.
Hasonló nehézségekkel küzd a Blender projekt, amely február elején kezdte meg az AI által készített javításokkal kapcsolatos irányelvek kidolgozását. Korábban a Linux Foundation, a Fedora, a GNOME, a Firefox, a Ghostty, a Servo és az LLVM fogadta el az ilyen esetekre vonatkozó szabályokat. Januárban a GitHub megkezdte a diskurzust a minőségellenőrzés nélküli, AI-asszisztensek által generált, alacsony minőségű pull-kérelmek korlátozásának bevezetéséről, amelyek nem felelnek meg a minőségi követelményeknek, és komoly nehézségeket okoznak a karbantartók számára.
Forrás: opennet.ru
