Does using an em dash make your writing look AI-generated? The question sounds simple, but authorship is harder to infer than a single punctuation choice might suggest. This episode explores why the em dash became associated with AI writing, what broader text patterns can show, and where those patterns stop being useful for judging an individual piece of writing. It also looks at the limits of human judgment and automated AI detectors, including the complications created by different models, different kinds of writing, prompting, editing, and varying degrees of AI involvement. The practical issue is not whether em dashes ever appear more often in AI-associated text, but how much weight one visible stylistic cue should carry when deciding who—or what—wrote something.
Listen to the episode — then explore the full transcript, the clarity poll, and evidence in The Evidence Edit.
