AI slop fatigue is turning into policy

The complaints show up in unrelated threads. The Llama-modchip README is called "obnoxious" and full of AI slop. The Claude Code leak write-up is called 80% repetition with a note to edit AI output before publishing. One commenter on the JIT bitmap text post says it is "a breath of fresh air to read something which is not written by an LLM." The older Ask HN to add a flag for AI-generated articles (48886741) shows the community wrestling with how to enforce this. It cites a new guideline, "Don't post generated text or AI-edited text," and the problem that detectors are unreliable.

The cost is credibility, not just taste. Readers now assume machine-written docs mean less careful engineering underneath. One commenter on the flag thread says the stigma got bad enough on Lobsters that they felt unwelcome even without using LLMs to write.

The nuance is that enforcement is hard. Voting can be gamed, and false accusations hurt people who write by hand.