ChatGPT is great—apart from the fact that everyone sounds like a robot. In this video I show you how to spot AI writing fast, then how to fix it with a handful of practical prompt hacks that actually work. We start with the obvious tells: the dreaded em dash (yes, that extra-long hyphen ChatGPT is weirdly obsessed with), corporate LinkedIn word-soup like “transformative” and “synergistic,” and formatting giveaways like huge line breaks, emoji-stuffed headings, and random bolding in bullet lists. Then I go deeper into the patterns that give AI away even when the vocabulary looks “fine”: the overused “it’s not just X, it’s Y” structure, the addiction to triplets (clarity, consistency, convenience), and the way it repeats itself—make the point, explain it, reword it, wrap it up. From there, I walk you through my workflow to humanize AI text: set custom instructions to ban the tells, prime every prompt with a short “write like a human” style guide (sharp, specific, imperfect, no filler), and finally run the output through a self-critique prompt that flags anything robotic and rewrites it. The goal isn’t to “sound human” as a gimmick—it’s to communicate like a real person without the AI stench.