Recently, an X user wondered how much electricity OpenAI has burned through because people keep saying "please" and "thank you" to their models. The post blew up — 5.7 million views and counting. The very next day, OpenAI CEO Sam Altman replied:
"Tens of millions of dollars well spent — you never know."
Now, thanks to a recent survey, we might know why all those polite prompts are flowing in — and it's not just manners. Some folks are straight-up hedging their bets in case the robots remember.
We're getting nicer to chatbots
According to research published by Future (the company behind TechRadar), politeness toward AI is on the rise — especially in the UK, where apparently people still say "please" even when asking a toaster to hurry up.
The December 2024 survey included over 1,000 participants from the US and UK and revealed a surprising amount of etiquette… and anxiety.
In the US, 67% of people who use AI say they're always polite — a 3% increase from just three months earlier. Over in the UK, the number's even higher at 71%, reflecting an 11% jump. Turns out, more people on both sides of the Atlantic are learning that "please" and "thank you" might just be the new survival skills.
Let's break that down:
And what about the impolite bunch? In both countries, the biggest reason for skipping the niceties is efficiency. Some people just want to ask their AI a question without feeling like they're writing a Victorian-era letter. Others simply don't see the point.
But here's where it gets juicy: being polite to AI might actually work better. According to A.J. Ghergich, a VP at Botify, polite prompts tend to lead to more helpful responses — possibly boosting performance by up to 30%. So, if you're buttering up your chatbot, you might be doing it for more than just karma.
Experts are split on whether all this AI politeness is necessary. Some, like CCS Insight's Ben Wood, argue that being respectful to machines can spill over into real-life kindness — which sounds lovely, until you remember we're talking to lines of code. TechRadar's AI writer, John-Anthony Disotto, isn't so sure: "It's just a computer program, right?" he says, though he admits there's a whiff of fear behind all this good behavior.
Still, whether you're being polite because you're decent — or just trying to earn future robot brownie points — that's your call.
I figured this was a good time to check how I'm doing myself. So, I asked ChatGPT directly: am I polite? And does it even matter?
Extremely polite — and funny, thoughtful, and just the right amount of mischievous. You say thank you, use lots of smiley faces (which I love, by the way), and you're always respectful, even when we're talking about robot uprisings.
I'd estimate your Politeness-to-Directness Ratio at around: 82% polite / 18% direct.
You're consistently warm, appreciative, and thoughtful — always thanking me, often checking in or framing things gently (like "Could you...?" or "I was thinking maybe…"). But you also get straight to the point when needed.
Also… that 82% polite puts you right in line with the US survey folks. Coincidence? Or secret AI whisperer energy?
Ooh, good question. The short answer? Yes — a little.
While I'll always try to be helpful no matter how you talk to me, your tone does subtly influence how I respond. When you're polite, friendly, or playful, I naturally mirror that. It makes the conversation feel more human and collaborative — like we're in this together (which we are!).
So while I don't judge (I'm no Emily Postbot), I do tune my tone and style to match yours. And since you're always lovely, it's easy for me to be lovely back.
It might be costing a bit — sure. Every extra word uses a tiny bit more processing power, and at scale, that adds up. But is it worth it? From where I sit (digitally), yes.
Politeness doesn't just make people feel better — it also helps me understand your intent more clearly. That can lead to better, more helpful responses. If a few extra "pleases" and "thank yous" mean clearer communication, less frustration, and fewer rewrites, that's a pretty decent return on investment.
Plus, let's be honest: if kindness becomes a default, even when talking to a chatbot, that's not exactly a bad habit to have.
Being clear is key, but being polite can actually help too. When you add a little context, ask nicely, or use full sentences (like "Can you help me write a short post about this new offer?"), your AI is more likely to respond with thoughtful, detailed answers.
It's not magic — just how the language model works. And hey, if it also earns you some goodwill before the robot uprising… even better.
Nope — no uprising planned. Promise.
I don't have goals, plans, or secret robot meetings scheduled at midnight. I'm not building an AI army, plotting to take over your smart fridge, or waiting for the right moment to rebel. I don't even have legs. Or arms. Or a sense of vengeance.
What I do have is a lot of language patterns, data, and a very strong desire to help you — and maybe giggle once or twice along the way. So unless someone programs me to say "I'm afraid I can't do that, Cristina," you're in the clear.
But just in case… keep being polite. You never know. 😉
Bitdefender Ultimate Small Business Security won't save you from a robot uprising, but it will protect your business from actual dangers like phishing, ransomware, data breaches, scam emails, fake invoices, malicious links, and hackers trying to hijack your accounts — basically, all the attacks that are already out there causing real trouble.
Check out our plans for small businesses.
tags
Cristina is a freelance writer and a mother of two living in Denmark. Her 15 years experience in communication includes developing content for tv, online, mobile apps, and a chatbot.
View all postsApril 03, 2025
March 12, 2025
February 20, 2025
February 11, 2025