Hi!
On Wed 2025-07-09 15:14:38, Shuah wrote:
On 7/8/25 14:39, Pavel Machek wrote:
Hi!
So... I'm afraid subject is pretty accurate. I assume there's actual human being called "Sasha Levin" somewhere, but I interact with him via email, and while some interactions may be by human, some are written by LLM but not clearly marked as such.
And that's not okay -- because LLMs lie, have no ethics, and no memory, so there's no point arguing with them. Its just wasting everyone's time. People are not very thrilled by 'Markus Elfring' on the lists, as he seems to ignore feedback, but at least that's actual human, not a damn LLM that interacts as human but then ignores everything.
You aren't talking to an LLM - My understanding is that Sasha is sending these patches (generated with LLM assist) and discussing them on mailing lists.
At this point, I'd like to know (a) what steps (if any) were taken to prevent LLM hallucinations from reaching the lists, and (b) what steps (if any) were taken to make sure patches Signed-off by developer were actually reviewed by said developer, and not applied simply due to said hallucinations.
Confusion caused LLM hallucinations can be seen for example in thread "[PATCH AUTOSEL 6.15 6/8] PM: Restrict swap use to later in the suspend sequence", it includes as great stuff as fake http links.
I have seen solution proposed for (a), but have not seen any solution proposed for (b) and that's actually more serious problem.
One solution would be to use separate email address "Autosel bot <>" for both From and Signed-off, so there's no confusion between content generated by developer and content generated by LLM.
Best regards, Pavel