When Your Employee's AI Argues With Your AI: What's Actually Left of the Conversation

A year or two ago, if an employee's email suddenly read like it had a research team behind it, you'd have noticed. You might have even been impressed. Today, you already know. I already know. We all know. And somehow, that hasn't stopped anyone from doing it.

That's the part I actually want to talk about. Not "your staff is using AI," because you already knew that. What I'm more interested in is what happens once everyone knows it's happening, nobody says it out loud, and the whole thing keeps going anyway.

The tell isn't AI. It's effort.

I've run the same practice with a lot of the same people for years now, so I know their writing. I know how they think, how they argue, what they sound like when they're frustrated versus when they've actually done their homework. So when an email shows up that doesn't quite sound like them, I notice. That part isn't new. What's new is the shape of what shows up instead.

Here's what I've actually learned from watching this play out in real raise conversations. When someone sends me something vague, all buzzwords, "improving throughput," "successfully leading initiatives," basically their job description dressed up in bigger words, I push back. Hard. And almost every time, the response is some version of "okay, that's fair," and they keep working for us the next year. No ultimatum, no drama. I called the bluff, and there wasn't much behind it.

But when someone sends me a well-built argument, real market comparisons, specific context about what they've actually taken on, numbers that hold up, I take it seriously. And here's the honest part: I do notice, somewhere in the back of my mind, that it might have taken them fifteen minutes with an AI tool instead of two hours alone at a laptop. And I've caught myself weighing that. But when the substance is real, it stays real regardless of how fast it got assembled. I've given raises off the back of exactly this kind of email. If the case holds up, the case holds up.

So the tell was never really "did AI touch this." The tell was always effort and substance, whether someone actually thought about their argument, or just asked something to generate the appearance of one. AI didn't invent that gap. It just made the low-effort version look a lot more polished than it used to.

You already knew this, and that's kind of the point

If none of that surprises you, good. It shouldn't. A third of workers admit they hide their AI use specifically because it gives them an edge over their peers. And honestly, I get it. I'm using AI to help draft this very post, because it lets me get a genuinely good idea out of my head and onto the page in a fraction of the time it would otherwise take me. That's not a confession. That's just a smarter way to work.

One global study found that most employees have avoided revealing when they use AI, and more than half have presented AI-assisted work as their own without flagging it. I don't read that as some grand deception. I read it as people quietly figuring out that if you can produce the same quality of thinking, the same argument, the same idea, in half the time, you'd be foolish not to. Life isn't only about work. If a tool gets you to a better outcome faster, that time belongs back to you, your family, whatever you'd rather be doing instead of staring at a blank draft for two hours.

Which raises the actual question I keep sitting with: if everyone knows, and everyone still does it, does it make the message any less real?

Here's where I've landed, mostly

My honest answer is that substance doesn't care who typed it. A strong argument is still a strong argument whether it came out of someone's own head over two hours or out of a prompt box over fifteen minutes. If the market data is accurate and the case is sound, it's sound. I'm not going to penalize someone for using a tool efficiently to make a true point. That would just be me punishing good reasoning because of how it arrived, which isn't actually a principle I want to hold as a manager.

But I'll be honest about the uncomfortable flip side too. When I get one of the vague, low-effort ones, the kind that's clearly just a job description run through a paraphraser, my instinct lately hasn't been to sit down and carefully draft a thoughtful, personal response. It's been to think, well, if you didn't put real effort into this, why am I obligated to? I've genuinely caught myself thinking about handing my own AI tool the exact same prompt and letting it draft the counter. That's the moment that actually unsettles me a little. Not the individual instance of it, but what it implies at scale. If both sides know the game, and both sides start playing it back, at some point the actual negotiation is happening between two AI tools, with two humans supervising from a distance and signing off on whatever comes out.

So where does that leave us?

I don't think I have a clean, tidy answer to this, and I'm not sure anyone fully does yet. I know substance should matter more than authorship, and I stand by that. I also know that once everyone's aware of the game, pretending nobody's playing it stops being realistic, and at some point choosing not to play starts to feel like showing up to a negotiation with one hand behind your back.

Maybe the honest version of where we're at is this. We're all quietly aware that a good chunk of what lands in our inbox had some help, we mostly don't say so, and we keep operating like it's fine. Maybe it is fine. Maybe substance-over-authorship really does hold up as a principle no matter how far this goes. Or maybe we're a few years away from realizing that once both sides fully hand the writing over, something in the relationship itself quietly changes, even if nobody can point to the exact moment it happened.

I don't know yet. But if you're running a practice and you haven't noticed this shift in what's landing in your own inbox, I'd bet you will soon. And when you do, it's worth asking yourself the same question I'm still sitting with: what does it actually mean for something to feel real, once everyone knows how easy it's become to fake the effort behind it.

Want to talk through what this actually means for how you're managing your team right now?