Slop anxiety: people want proof a human did the work

Reaction to OpenAI's withdrawn math results leaned on one word, slop. The same instinct shows up elsewhere in the batch. An Ask HN asks for a flag for AI-generated articles, with replies saying voting is easy to game and AI detectors aren't reliable. Someone quoted the HN guideline: "Don't post generated text or AI-edited text. HN is for conversation between humans." Another thread had Cliff Stoll saying "rumors of my death are slightly exaggerated" after an AI-written obituary, and a commenter worried about AI slop being the easy way to grow accounts.

Even the fun posts get the same question. On the Quake port to safe Rust, the first reply was "How? Was it a LLM-assisted rewrite?" The new default assumption is that output may be machine-made and unverified, so provenance becomes part of the product.

A real counterpoint is that policing AI authorship risks stigma against people who use tools well, and detection can't be done reliably. Commenters admitted the line is hard to draw.