GCC steering committee announces AI policy
Points and comments are a snapshot, not live.
The GCC steering committee has banned AI-generated code in legally significant contributions.
The GCC steering committee announced a policy barring legally significant contributions (over ~15 lines) containing LLM-generated content. The policy, recommended by a working group, uses the GNU project's definition of legal significance. Maintainers may accept AI-generated test cases. The policy does not restrict LLM use for research, analysis, or bug discovery, as long as output is not included in contributions. The committee expects the policy to evolve and will revisit it periodically.
What commenters are saying
Commenters split sharply on enforceability and rationale. One camp argues the policy is unenforceable and will breed covert AI use, calling it an ideological attack on a tool that is here to stay. Another counters that AI contributions are often low-quality slop, that reviewers can detect them consistently, and that project maintainers have the right to set quality standards. A third camp notes the policy is actually permissive-allowing AI for tests and small patches-and that similar debates are occurring in LLVM.