They're pretty bad about fixing issues it seems. My favorite is #2472 which they demonstrated "fixing" on stage on the release of GPT 5, but the ticket is still open and the "fix" hasn't been merged. The original blog that flagged this fact https://blog.tymscar.com/posts/openaiunmergeddemo/ and the issue: https://github.com/openai/openai-python/issues/2472
Claude meanwhile just auto closes all issues because they simply don’t triage ~anything. There’s been countless instances of this horrifying issue created for over a year now: https://github.com/anthropics/claude-code/issues/16180
> Permission bypass when commands are chained with &&
At one point they fixed their auto stale bot closing bugs but, hey, guess that wasn’t long lived.