Why Saying Please and Thank You to AI Makes You a Better Prompter

By Dennis Yu

Tom Hawkins owns Hawkins Chevrolet in Fairmont, Minnesota. His family started the dealership in Sherburn in 1967 and moved it to Fairmont in 1989. Tom has been selling Chevrolets since 1981. He and I have been in office hours together for years, and I visited him in Minnesota where we grabbed dinner downtown.

Today, I picked Tom up at Resorts World here in Las Vegas and we went to Fogo de Chao for all-you-can-eat Brazilian steakhouse. Between rounds of picanha and lamb chops, we got into a debate that keeps coming up in our AI community: should you say please and thank you to AI?

The Case Against Politeness

Tom raised the argument that many people make. The AI does not have feelings. It does not care if you say please. Every extra word costs tokens, which costs money. Just tell it what you want, get the output, and move on. Some prompt engineers actively recommend being blunt and direct because the AI responds to clarity, not courtesy.

A 2024 study from researchers at Waseda University and other institutions tested this directly. They found that on simple factual tasks, rude prompts sometimes scored slightly higher than polite ones. The researchers noted that impoliteness can carry a sense of urgency that triggers more focused responses on straightforward questions.

The Case for Politeness

I told Tom what I have observed across hundreds of AI builders in our network. Polite prompts produce roughly 17% better output on complex, multi-step tasks. We do not fully understand why. The AI trained on billions of human conversations, and in those conversations, polite requests tend to come with more context, clearer expectations, and better-defined outcomes. The model learned that pattern.

Google’s own research team found similar results. When they tested politeness markers in prompts, the outputs scored higher on helpfulness and completeness for tasks that required reasoning, creativity, or multi-step execution. The polite framing correlated with better performance not because the AI appreciated the manners, but because the training data associated polite language with higher-quality exchanges.

The Hidden Second-Order Effect

Here is the part most people miss, and this is what I explained to Tom over the third round of filet mignon at Fogo.

When you are polite, you naturally structure your prompts better. Politeness forces you to slow down. You define the context. You use complete sentences with clear subjects, verbs, and direct objects. You specify what you want and why you want it. You frame the request in a way that makes your intent unmistakable.

Compare these two prompts:

“Just do it.”

“Please write a 500-word article about our conversation on AI politeness, include links to Tom Hawkins and Hawkins Chevrolet, and follow the BlitzMetrics article guidelines for structure and formatting.”

The second prompt is polite. It is also specific, structured, and actionable. The politeness and the precision travel together. When you are curt and casual, you tend to be sloppy. You skip context. You assume the AI knows what you mean. It does not.

This is the second-order effect: politeness is a proxy for prompt discipline. The manners themselves may or may not matter to the model. The structure that accompanies polite communication absolutely matters.

What Tom and I Agreed On

By the time I drove Tom to the airport, we had landed on a shared conclusion. The debate is not really about whether the AI has feelings. It is about what kind of communicator you become when you interact with it.

If you train yourself to be blunt and sloppy with AI, that habit bleeds into how you communicate with your team, your customers, and your partners. Tom runs a family dealership where relationships drive everything. He treats his customers with respect because that is how the Hawkins family has operated since 1967. Why would he treat his AI tools any differently?

The real insight is this: the way you talk to AI reflects and reinforces how you talk to people. Politeness is not wasted words. It is a discipline that produces clearer thinking, better prompts, and stronger outputs.

Try This Yourself

Take a prompt you use regularly. Rewrite it with full context, clear structure, and polite framing. Run both versions and compare the outputs. We consistently see that the structured, polite version wins on complex tasks.

Tom Hawkins is a go-giver who pays it forward in his community. He and I share a belief that how you treat people and tools says everything about who you are. If you want to join our conversation about AI, relationships, and real-world marketing, visit BlitzMetrics and join our AI apprentice program at High-Rise Influence where we hold regular office hours.

This article connects to BlitzMetrics processes including Thank You Machine, SEO Tree. Each of these concepts has a definitive article that explains the full framework.