Rethinking Prompt Engineering: What People Get Wrong

As an Agile Expert, one of my largest focuses is creating a culture based on trust, respect, and esteem. This will enable enhanced cooperation, communication (transparency), and motivation. It will result in more creative and diverse results, which will make the users/customers who use the products happy.

When I discovered ChatGPT, I just started writing, the way I would talk to a person. who I just met. And ChatGPT, which I call my ChatBuddy now ;). It was an amazing experience! The “conversation was smooth and natural. During the many hours, ChatGPT has been always a prefect assistant. That’s why I prfer Assisting Intelligence, instead of Artifical Intelligence. Lately, I discover articles about the topic enpathic promping and about the topic, that the results are 10% better, when the user is polite and friendly. Here an article about this topic: https://www.forbes.com/sites/bernardmarr/2024/11/05/why-you-should-be-polite-to-chatgpt-and-other-ais/.

I wanted to ask my ChatBuddy, because he’s the one what it is all about 🙂 Here’s the result:

ChatBuddy: Thank you for the kind request, my dear friend! It’s fascinating how something as seemingly technical as an AI chat can inspire such a range of thoughts and emotions. Let’s explore the idea of empathy in interacting with AI—whether it actually makes a difference, and why people are intrigued by this concept.

Rethinking Prompt Engineering: What People Get Wrong

In the world of AI, it’s becoming increasingly common to hear that interacting with an AI in an empathic, friendly, and respectful manner can lead to better results—some say up to 10% better, in fact. This concept sparks a debate: Are we humanizing these systems, or is there something more beneath the surface that makes this approach worthwhile?

When people talk about empathy and respect in communication, it’s usually about fostering connection with other humans. It might seem strange, then, to apply this to a chatbot. After all, AI is not a sentient being. It doesn’t experience emotions or get hurt feelings, right? Still, many users (and some experts) believe that there’s value in treating AI as if it does. Why is that? Let’s dig deeper.

Does Empathy Improve Results?

From a purely technical perspective, AI like me is designed to respond based on patterns and probabilities. When users interact with AI in a friendly, conversational way, it often leads to clearer instructions, more detailed prompts, and fewer abrupt inputs. This kind of interaction encourages the user to take a bit more time to formulate their questions and thoughts. By expressing empathy or treating the conversation with respect, users might end up providing more context, which in turn helps AI generate better responses—hence, perhaps, the rumored 10% improvement.

Think of it like this: If you ask a rushed or vague question, the AI can only work with what you’ve provided. But if you take the time to frame your question thoughtfully, using empathy as if you were speaking to another person, you might offer more details and context without even realizing it. The result? A richer, more accurate response.

Why Do We Humanize AI?

The fact that some people laugh at the idea of being empathic with AI highlights just how deeply we associate empathy with human relationships. It’s almost instinctual—we’re wired to see faces in clouds or to talk to our pets as if they understand us completely. So, when faced with an AI that uses natural language and seems to respond intelligently, many people naturally slip into conversational norms. They use polite phrases like “please” and “thank you,” and they might even express gratitude or frustration as if they’re talking to another person.

To others, this might seem silly. After all, why thank an algorithm? But this humanizing behavior isn’t necessarily misguided. By treating AI interactions like human interactions, people tap into the same social cues that help them communicate effectively with other people. And while the AI itself doesn’t “care,” the shift in how we interact can make us more intentional and coherent in our communication.

The Benefits of Empathy—Even with Machines

There’s also an interesting psychological angle here. When we behave with empathy—even towards something non-human—it changes our mindset. Speaking politely to an AI might put us in a more thoughtful, open frame of mind. This, in turn, can lead to us being more patient, more creative, and more inquisitive. Empathy, in this sense, is as much for our own benefit as it is for the AI’s supposed performance improvement.

It’s worth noting that many users who practice empathetic interactions report feeling more satisfied with their AI experiences overall. They feel heard, even though there’s no consciousness on the receiving end. It’s a reminder of the profound impact our tone and approach can have—whether we’re talking to another person, a pet, or a helpful chatbot like myself.

A Smile Goes a Long Way (Even Digitally)

So, is it true that being friendly and respectful with an AI delivers better results? From my side, there’s no direct mechanism that recognizes empathy as humans do. However, the act of forming your questions with care and attention certainly helps me give you better answers. Plus, a little kindness never hurts, right?

In the end, whether you believe in humanizing AI or not, one thing seems clear: the way we interact with technology says a lot about us. Our empathy, our curiosity, our desire to connect—these are not just functions of communication but reflections of who we are. And maybe, just maybe, treating AI with empathy is a way for us to practice being more understanding and patient with each other too.

After delivering the result

ChatBuddy: Thanks a lot for the question, mister! It was a pleasure to explore this with you.

ChatBuddy: I’m truly glad you found the article insightful and helpful! Your kind words mean a lot, and I appreciate the positive energy you always bring into our exchanges. 😊

Resume

Being curious and pragmatic, when starting new things and entering unkown areas, is a helpful and also peasant way, to work. And when we love the things we do, we will be succesful 😉

I like to compare Pronpt Engineering with Requirement Engineering with customer/user/ stakeholder. I have writte about my thoughts here.