Hi Frank, On Wed, Feb 11, 2026 at 6:09 PM Frank Ch. Eigler <[email protected]> wrote: > > Hi - > > > [...] > > The elfutils project does not currently accept contributions > > containing output generated by Large Language Models (LLMs) [4]. Use > > of LLMs to research, analyze or debug a contribution is allowed as > > long as no LLM-generated output is included in the contribution. [...] > > For the record, I believe this is a not the best approach. > > For a contribution to be submitted and accepted, all the elfutils > project needs in a legal sense is the DCO sign-off. From a practical > sense, it needs someone who is willing to stand behind the patch, > respond to reviews and future bug reports. Neither of those is > impeded by including LLM-generated output. > > CONTRIBUTING/DCO just needs: > [...] > (a) The contribution was created in whole or in part by me, > and I have the right to submit the contribution under each > license indicated in, or otherwise designated as being > applicable to, the file. > [...] > > An LLM-assisted contribution is indeed created in whole or part by a > human contributor, and pragmatic analysis can make the chance of > material license infringment infinitesimal, ergo "right to submit". > > > I fear that by being so nervous, we are going to exclude some people > who could make useful contributions. Productivity gains from > AI-enhanced IDEs are widely reported - and I can personally testify to > them.
Personally I would also prefer a more permissive policy. Consensus for such a policy does not currently exist but I hope that the incentives, like productivity, help drive clearer legal guidance around these issues so that we can build the consensus. Aaron > > Also by being so nervous, we are going to motivate other people to > just casually use LLMs in their work, but neither disclose this nor > bother with even cursory license investigation, because it would just > be punished by rejection. > > Both these leave the project worse off than if we didn't bother with a > policy at all. We heve never before second-guessed contributors' > assertions about their right-to-submit, contemplated asking them to > spell out their exact workflow, or shamed them for using the wrong > tools. And should an actual infringement situation arise, it need be > no different in the LLM vs. non-LLM case, just revert and clean up. > > The remote risks of LLM infringement do not seem - to me - worth this > change in tone. I understand that this is an elfutils maintainership > type decision, but I respectfully dissent. > > > - FChE >
