Are you nice to AI? (And does it even matter?)

Here's a fun (and slightly weird) question: Are you polite when you talk to AI?

It sounds like a silly thing to ask, after all, we're talking to a machine, right? But here's a twist. The way you speak to AI can change how it responds to you.

And no, it's not because the AI has feelings (although it can feel that way sometimes). It's because of how these systems are trained and how much they mirror the way we interact with them.

Researchers tested this, and the results are fascinating

In one study, users who were polite and respectful, using phrases like "please" or "can you help me with..." got longer, more accurate, and more thoughtful answers from AI.

But when the users were abrupt, aggressive, or demanding, the quality of the AI's responses dropped. The answers got shorter. Less helpful. Sometimes even less accurate.

So what's going on here?

AI doesn't have feelings, but it does reflect patterns

Most modern AI tools are trained on billions of conversations, messages, articles, and forums. That means they've picked up not just information, but the patterns of how people talk.

When you speak to an AI, it doesn't just try to answer your question, if often tries to match the tone you're using. So if you're patient, it's more likely to be patient back. If you're polite, it mirrors that. And if you're harsh or rushed, it might cut corners too. It's like holding up a conversation mirror and that has some interesting consequences.

So...are we training AI, or is it training us?

That's the big question, isn't it?

On one hand, the way we talk to AI gives it feedback. We reinforce certain tones, styles, and behaviours every time we use it. But on the other hand, it also influences us. The more we interact with AI that rewards calm, clear communication...the more we might start communicating that way ourselves.

I don't think that's a bad thing. In fact, it might be a gentle reminder that the way we speak, whether to a human or a machine matters more than we realise.

What this means for teams and workplaces

Teach AI etiquette. It's not about being overly formal. It's about being intentional with your language. Clear and respectful prompts tend to produce better results.

Model good prompting behaviour. If your team uses AI tools regularly, encourage them to take an extra second to frame their input with clarity and tone in mind. It makes a difference.

Notice the feedback loop. If a tool "feels" frustrating or unhelpful, check the tone you're using. You might be surprised how much it changes when you change your approach.

AI doesn't care how you speak to it, but it does respond differently depending on how you do. The next time you're prompting ChatGPT, Claude, or any other tool, try this simple test: Ask the same thing two ways, once politely, once abruptly, and see what happens.

You might just start treating your AI like a colleague and getting much better results in return.

Do you think the way we speak to AI is shaping how we work and communicate? Or is it just a reflection of us?