Being short with AI can help the planet

Being short with AI can help the planet

We have all been taught to be polite: say “please,” “thank you,” even to chatbots. But what if those friendly words to ChatGPT are quietly increasing your carbon footprint?

Yes, you read that right. Being too nice to AI might be bad for the planet.

Let’s dig into why doing small talk might be smarter — and greener.

The Politeness Price Tag

Sam Altman, CEO of OpenAI (aka the boss of ChatGPT), recently mentioned that the simple act of adding words like “please” and “thanks” could be costing tens of millions of dollars every year extra in computing costs.

Why? Because every message you type is broken down into tokens — small parts AI uses to understand you. The more tokens, the more work the AI has to do. More work = more electricity = higher costs and more environmental strain.

With over 350 million users each week and a billion prompts daily, even a few extra words from everyone add up to a big environmental toll. It’s like the energy version of death by a thousand paper cuts.

Short and Sweet = Energy Efficient

Want to go green while using AI? Keep your messages short.

Example:

●       “List steps to install Python.” uses less energy than

●       “Hi ChatGPT, could you kindly list the steps to install Python for me? Thank you so much!”

It’s not that politeness is bad but in an environment where millions of users are asking questions every second, efficiency becomes key.

But Wait — Politeness Has Its Perks

Before you start firing off one-word commands to your chatbot, pause for a moment. Being polite can actually result in better responses. According to AI expert Dr. Lance B. Elliot, polite prompts can result in more helpful, detailed, and accurate replies. This is because AI is trained on human conversations, and polite ones usually have better structure and clarity.

So:

●       Being polite = Better results

●       Being brief = Greener use

It’s a classic “choose your own adventure” scenario. Go green or go deep.

What Are People Actually Doing?

A recent survey by Future PLC gave us some interesting stats:

●       70% of users speak politely to AI tools like ChatGPT.

●       Only 12% do it because they think it might keep them safe during a robot uprising (seriously).

The rest? They're just being polite because it feels natural, or because yelling at your digital assistant feels a bit weird. Whether it’s habit or hope for good karma, most users are choosing the friendly route.

 

Final Thoughts

AI doesn’t have feelings (yet). ChatGPT won’t judge your tone, get sad if you skip the “please,” or report you to robot HR.

But energy usage? That’s real. And every word you add increases it a little.

So next time you are writing to ChatGPT:

●       Keep it short to save energy

●       Be polite if you want top-notch answers

Either way, now you know how something as small as a “thank you” can have a bigger impact than you thought.

This opens up an interesting question: What if AI did start judging our manners? Imagine getting a message from ChatGPT saying, "I'm sorry, your request is being delayed because you forgot to say please. Try again!"

For now, though, ChatGPT is too busy crunching numbers to care about your tone. But who knows what the future holds?

P.S. If robots ever take over, I will still be saying “please” — just in case they give out snacks to the nice ones :(

To view or add a comment, sign in

Others also viewed

Explore topics