AI-generated contributions are overwhelming open source maintainers with low-quality PRs, fake security reports, and unreviewed agentic output. Projects like curl have had to suspend their bug bounty programs and even halt vulnerability reports for a month. Many major open source projects (CPython, Django, Kubernetes, Linux Kernel, NumPy, etc.) have adopted AI contribution policies that allow AI-assisted work as long as a human reviews and owns the result, while banning fully automated submissions. PostgreSQL notably lacks any such official policy, which leaves its broader ecosystem of extensions and tools without guidance. The author calls on the PostgreSQL community to develop responsible AI usage policies, framing the issue as one of respect for maintainers' time.
