I think we're slowly seeing the severe implications of AI coding slop creeping up. In the last 24 hours we have received a bunch of bug reports (against non-upstream kernels even) on fsdevel that are almost guaranteed to be generated by some bot.
In addition we seem to be getting really strange security@kernel.org reports that report normal system behavior as security vulnerabilities.
I can't wait for the next 6 months...
@brauner in the last two weeks we had at least 4 ""security"" bug reports in the systemd bug bounty program that were clearly llm-generated slop, down to hallucinated non-existing commands in "reproducers", and obvious nonsense in the description with placeholder and non-existing CVE numbers, quoting: "This document provides comprehensive technical evidence demonstrating that CVE-2023-XXXXX is a critical security vulnerability in systemd-boot that must be addressed."