Someone reads something you wrote — an email, a proposal, a post you were a little proud of — and they tell you it sounds like a machine wrote it. Not "this could be tighter." Not "strong draft, a few notes." They say it sounds AI-generated. And every instinct in your body fires the same four words: but I wrote it.

That reaction is universal. It is also, without exception, the wrong move. Not because the feeling is wrong — the sting is real, and we'll get to why — but because those four words close the one conversation that could actually help you. This is what Layne Robinson unpacks in the latest episode of Managing A Career: the half-second after you receive feedback that feels factually wrong, and what you do in that moment that decides whether the feedback helps your career or quietly damages it.

The distinction that changes everything. When someone gives you feedback, they are almost never handing you a fact you can disprove. They are handing you a perception. And a perception is not a verdict you can appeal — it is a report on how you landed. "Your post sounds AI-generated" might be completely false as a statement of authorship — you wrote every word at your kitchen table — and still be completely accurate as a perception. Because it is true that they read it and felt a machine on the other end. You can win the argument about the fact. You will lose the thing the fact was pointing at.

This perception-versus-verdict distinction is the spine of the episode. Layne traces it through a story that isn't his — a LinkedIn post from writer and coach Khushi Lulla, who shared that one of her clients called her own writing AI-generated. A professional writer. Her own words. Called artificial by the person she was trying to serve. What Lulla did next is the lesson: she didn't fire back, didn't pull up her drafts to prove authorship. She kept reading the feedback — she stayed in the discomfort — because she wanted to understand why the client saw it that way.

Layne commented on that post because he had been having nearly the same conversation with people on his own team. Different words, same shape. Someone hears something about their work that they are convinced is simply not true, and they spend all their energy proving it isn't true. Over thirty years, he has watched how rarely that works.

Reacting is not responding. The episode draws a hard line between two things that feel identical from the inside but produce completely different outcomes. Reacting is the four words. But I wrote it. It is instant, defensive, and aimed at protecting you. Responding is what Khushi Lulla did — staying in the discomfort long enough to get curious about where the perception came from. One closes the conversation. The other opens it.

Curiosity here is not a soft skill or a personality trait. It is a tactical choice. As explored in a Psychology Today piece on criticism and defensiveness, you can be genuinely curious about a perspective you think is dead wrong without agreeing with a word of it. Curiosity is not surrender. You are not conceding the point. You are collecting information you cannot get any other way.

The person giving you feedback is not a judge. They are a witness. They are describing what they saw from where they were standing. And a witness who feels attacked stops talking. A witness who feels heard tells you everything — including the part that actually helps you. When you feel those four words rising, that is your signal. Not to speak. To listen harder.

Why it stings so much. Layne is honest about the difficulty. "It sounds AI-generated" does not just inform you — it stings. The episode draws on a framework from the book Thanks for the Feedback by Douglas Stone and Sheila Heen, which lays out three triggers that make feedback hard to hear. Truth triggers — when we think the content is just wrong. Relationship triggers — when it is who said it that sets us off. And identity triggers — when the feedback pokes at our sense of who we are.

"Your writing sounds like a robot" hits two of those triggers simultaneously. It is a truth trigger, because you know you wrote it, so the content feels false. And it is an identity trigger, because your writing is you. Being told your voice sounds artificial is not a note on a deliverable. It feels like a note on your humanity.

That double hit is exactly why people react instead of respond. The sting is real. But naming it gives you a half-second of control. When you can say to yourself, okay, that's the identity trigger talking, you have created just enough distance to choose the response instead of firing the reaction. The skill is not "stop feeling the sting." You are going to feel it. The skill is feeling it and not letting it drive.

Decoding the signal underneath. Once you have paused and stayed curious instead of defensive, the real work begins. You have to decode the perception into the signal underneath it. Because "it sounds AI-generated" is a symptom, not a diagnosis. Nobody can act on it as stated. You have to dig for what they actually experienced. And when you do — when you ask "what specifically gave you that impression?" — it almost always resolves into something concrete and fixable.

It usually means one of a few things. It means the writing was too clean — every sentence the same length, every edge sanded off, no rhythm. It means there was no point of view — it summarized, it hedged, it never said I think or I disagree. It means there were no specifics — no real example, no number, no moment that could only have come from you. The structure was so balanced and so generic that it could have been about anyone, written by anyone, or anything.

The uncomfortable part: the things that now read as "a machine wrote this" are the exact things professionals were once praised for. Clean structure. Professional polish. No rough edges. For years those were the markers of competence. Now they are the markers of absence.

This connects directly to a broader shift Layne explored in AI is Eroding the Signals Employers Use to Judge Talent (MAC-142). The whole basis on which people judge your work is shifting. The feedback "this sounds AI-generated" is the early-warning siren. It is not an insult. It is a perception telling you that your fingerprints have worn off your own work. And once you have decoded it that far, the fix is obvious — and it is not "polish it more." More polish is what got you flagged. The fix is to put yourself back in. A specific story. A genuine opinion. A sentence only you would write. You do not make it cleaner. You make it yours.

This is bigger than one post. Layne scales the lesson beyond writing and beyond one comment. You do not get to control the story people tell about you. You do not get to walk into the room and announce "I'm strategic" or "I'm a strong writer" and have it stick. The narrative is built out of perceptions — a hundred small impressions, formed in rooms you are not in. What you do get to control are the inputs. And the only way to change an input is to first understand what people are actually perceiving, and why.

This is the same muscle explored in Put Yourself In Their Shoes (MAC-073) and in Acting on Feedback (MAC-074). This episode is the layer underneath both — the step that happens before you act, where you correctly read the perception instead of arguing with it. And it connects to Receiving Effective Feedback (MAC-012), one of the very first things ever covered on the show. The throughline across all of them: the feedback is rarely the point. The perception behind it always is.

Think about how this scales. "You're not seen as strategic." That is not a fact to dispute — your boss is not going to be argued into seeing you differently. It is a perception to decode. What are they watching you do that reads as tactical? "Your team seems junior." Same thing. "I can't really tell what you do all day." Same thing. Every one of those is a witness statement. Every one of them is decodable. And not one of them responds to but that's not true.

The action plan. Layne closes with four concrete steps any listener can apply immediately:

1. Pause Before You Defend. The next time feedback lands wrong, do not say the four words. Buy yourself time out loud: "Let me sit with that — can you tell me more?" That single sentence converts a reaction into a response, and it keeps your witness talking.

2. Treat the Person as a Witness, Not a Judge. Ask for the specific moment. "What specifically gave you that impression?" You are not asking them to justify a verdict. You are asking a witness to describe what they saw, so you ca...