)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":24245,"name":"Harald Jensås","email":"hjensas@redhat.com","username":"harald.jensas"},"change_message_id":"bc522ae0e8eb86e15a2090533db74a99739412b5","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"1fc71080_05cb70d4","updated":"2026-06-29 11:11:17.000000000","message":"I guess I triggered this with the heat-driver changes that I pushed a while back. In general I agree with the content, humans should be in the loop and humans should understand the code proposed. Monty said something that resonate with me in https://www.youtube.com/live/b_Q_Hp6-QPQ?si\u003d7Spsha2L5dUVqg9j regarding AI agents enabling us to implement ideas that we\u0027ve had for a long time, but did not have time to do. Pushing PoC code for early review feedback is also good IMO, essentially I wish the wording here was a bit more aligned with openinfra\u0027s policy \"provide a starting point, which is then reworked by the human author.\"\n\nI think Zuul as a project should follow up with requiring Signed-Off in proposed changes. I am not a legal expert, but requiring that reduces the projects risk, creates a \"clearer\" line of responsibility - e.g. pushes the copy-riught responsibility to the author proposing the code to the review system.","commit_id":"a50ebde078844c5558d9a3404db62cbc77e68669"}]}