A Vote About Bots Is Really a Vote About Trust

Debian is asking its developers how to deal with AI-assisted contributions, with choices ranging from allowing them freely to banning them completely. While the headlines focus on the philosophical debate around whether generated code is genuine, I see the real issue as one of trust. The supply of patches has always been limited by the reviewers who have the knowledge and time to assess them.

AI tools make it much quicker to produce patches, but harder to verify their quality. This imbalance threatens the project's ability to maintain its high standards. A ban may seem like a simple solution, but it is difficult to enforce, and outright bans could actually hurt trust by discouraging disclosure of AI assistance. Instead, accountability matters more than provenance.

Projects should ask whether contributors fully understand the code they submit and will stand by it if it fails. Disclosure norms are valuable even if unenforceable because they signal to reviewers that a contributor is taking responsibility for their work. Debian's trust is built on the reputation of its volunteer maintainers who have shown over time that they can be counted on.

The real question is whether this social fabric can survive the influx of well-crafted contributions from contributors with no prior investment in the project. The vote should not just determine policy for AI, but also define the responsibilities of maintainers and contributors to preserve the community's integrity.