Debian votes to let contributors code with AI

The Debian community has voted to permit contributors to utilize generative AI tools while creating their contributions, provided that developers maintain responsibility for code quality. The Linux distribution's community recently established a policy on AI-assisted coding tools and invited participants to cast votes on one of eight proposals, including an outright ban, cautious use, or avoiding LLMs due to their environmental impact.

Out of the eight proposals, proposal E – "Responsible Use of Generative AI" – emerged victorious. This proposal maintains neutrality regarding the use of generative AI tools in Debian's development, maintenance, documentation, and other media. It asserts that the community neither endorses nor prohibits the use of such tools, stressing that they can significantly enhance productivity when utilized responsibly, allowing volunteers to focus more on tasks requiring technical expertise, judgment, review, and collaboration.

The proposal also emphasizes that AI errors should not serve as an excuse for subpar contributions, reminding contributors that they remain accountable for the quality, correctness, maintainability, and legal compliance of their submissions, regardless of the tools employed in their creation. Furthermore, the proposal encourages contributors to disclose AI-assisted contributions, although it is not mandatory.

In contrast, the Gentoo Linux team has imposed a ban on AI usage, while NetBSD and OpenBSD prefer standard code contributions. Linus Torvalds, a key figure in the open-source community, has expressed openness to AI-generated contributions. In July, he explicitly stated that Linux doesn't belong to the anti-AI camp, as demonstrated by his utilization of AI to rectify a challenging bug.

However, Torvalds has also raised concerns about AI-generated bug reports, which can sometimes overwhelm the Linux security mailing list.