The AI Pink Tax: Why Women Apologise for Using AI (and Men Don't)
“Sorry, I used AI for the first draft of this caption, don’t hate me.”
Read that sentence again. Then try to picture the man in your mastermind writing it. He doesn’t. He never has. The sentence doesn’t exist in his drafts folder, because it never occurred to him that it should.
I think the AI pink tax is real, and women are the only ones paying it. Not in dollars this time. In apologies.
The same tool costs women more
Pink tax used to mean the same razor, pink instead of blue, two dollars more. A surcharge for existing in the women’s aisle. We named it, we got angry about it, and most of us learnt to buy the blue one out of spite.
The new version doesn’t show up on a receipt. It’s the same tool. The same ChatGPT tab open on both laptops. She adds three sentences of guilt before she posts. He posts.
She spends another three hours editing the email so it “doesn’t sound like AI”. He gives it a skim and sends it. (I’m painting a scene here, not quoting anyone’s inbox, but if you’ve spent time in founder communities lately, you’ve watched some version of it happen.)
The cost isn’t the subscription. It’s the hours. It’s the confession drafted and redrafted before the work is allowed out the door. It’s the low hum of “am I cheating?” running under every task the tool touches. That hum is expensive. It taxes the exact resource the tool was supposed to give back: your attention.
And the tax lands almost entirely on one side of the aisle. Women are more conscientious about AI use in business, and it’s not hard to see why. We’re the ones who get called lazy when we take a shortcut. We’re the ones whose warmth gets audited. We built businesses on connection, so any tool that touches the connection feels like a threat to the whole thing.
If you’re carrying that guilt, you’ve probably explained it to yourself the way most of us do: the guilt means I care. If using AI felt fine, that would mean I’d stopped caring about quality, about connection, about the people on the other end.
That read is wrong. But not for the reason you’d expect.
The fear isn’t irrational
Here’s where I’m supposed to tell you the guilt is silly and the robots are friendly. I won’t, because you’ve seen the receipts.
You’ve messaged a business with your credit card in your hand and had a bot answer a question you didn’t ask, hand you a freebie you didn’t want, and lose the sale on behalf of a human who never knew you were there. I’ve written a whole piece on where automation loses the sale, and the short version is: the moment a buyer reaches out ready and gets a canned response is the moment trust dies. Not slowly. On the spot.
You’ve watched an auto-DM system fire off two thousand responses and bury the one question that mattered under twenty pieces of noise. You’ve received the “Hey babe!” message that was sent to four hundred people and felt what it feels like to be a row in someone’s spreadsheet.
So no, the caution isn’t paranoia. It’s pattern recognition. Bad automation has genuinely burned people, and heart-led founders are right to flinch at anything that promises to “handle” their conversations. The relationship is the business. Protecting it isn’t precious. It’s the job.
Sit with that for a second, because I don’t want to reframe the fear away. The fear is earned.
The mistake isn’t being careful. The mistake is aiming the caution at the entire category, when only one use of AI deserves it.
The machine was never supposed to answer her
Here’s what nobody tells you: the machine was never supposed to answer the buyer. It was supposed to find her.
Those are two different jobs, and everything about the guilt collapses once you separate them.
Answering is the sacred part. Your voice, your warmth, the moment a real person asks a real question and you meet her there. That’s the part that made your business heart-led in the first place, and it should never be outsourced. An AI that writes your replies for you is the automation you’re right to be scared of.
Finding is the other job. Somebody has to read the inbox. Somebody has to notice that one woman has watched every story for three weeks and never commented. Somebody has to spot the difference between “how much is it?” asked out of curiosity and “how much is it?” asked with a card in hand.
I spent years doing that job by hand. Former therapist, 25,000 DMs analysed, and here’s the pattern that never budges: about 2% of any inbox is moving toward a yes right this minute, and nearly 80% of a seller’s DM energy goes to people who were never going to buy. The finding work is real work. It’s also the least romantic work in your business. Nobody fell in love with scrolling.
That division of labour is the entire reason ForesightHQ® exists. I built software that reads the inbox and points: her, right there. It doesn’t speak to anyone. It doesn’t send the freebie. It hands you a short list and gets out of the way, because what AI can see in your conversations was never the problem. What it says on your behalf was.
You didn’t start your business to sort a spreadsheet of feelings by hand. You started it to talk to the person on the other end once you knew she was ready.
What changes when you stop apologising
Delete the apology sentence. Not because you’ve stopped caring, but because the confession was never protecting anyone.
You don’t owe the algorithm a confession, and you don’t owe your audience one either. The people reading your caption can’t tell which draft was first. What they can tell, immediately and every time, is whether the DM they sent you got a human answer.
So move the caution where it belongs. Here’s the test I’d hold any tool to: does it talk to my people, or does it tell me where to look? If it talks to your people, be as suspicious as you like. That one can cost you buyers. If it tells you where to look, it isn’t touching the warmth at all. It’s returning the hours you’ve been spending on the unromantic part, so the romantic part gets you at full strength.
Picture the week that follows. The three hours of guilt-editing go back into two real conversations with the two people who were already moving toward you. The caption goes out without a confession stapled to it. The buyer who watched every story finally gets noticed, and the noticing is yours, in your own words, the way it always was.
The warmth was never outsourced. The machine never spoke.
The buyer doesn’t need your guilt. She needs your attention.
Back to that sentence in your drafts folder. “Sorry, I used AI, don’t hate me.”
The woman on the other side of your inbox has never once needed you to feel guilty. She needs you to notice her. She’s been signalling for weeks in a language most sellers were never taught to read, and every hour the tax takes from you is an hour she waits.
Drop the apology. Keep the warmth. Learn to read.
Your next client probably already messaged you. The 5 Signals Guide teaches you how to find them.
Get the Free Guide at kelseysilver.com/5-signals
And if you want to practise the reading live: Cycle 1 of the 14-Day DM Sales Sprint Challenge kicked off this morning, and the September round’s list is already open. Details here.
Get the 5 Signals — Free
Learn the buyer behaviors you're probably missing right now. No fluff, no pitch — just the framework.