LocationsHOW TO ASK AI THE RIGHT QUESTIONS—AND WHAT IT TEACHES US ABOUT OURSELVES AND OUR FUTURE
January 7, 2025
In May 2024, I decided to engage with artificial intelligence (AI) in a new way—like a conversation with a mentor over coffee. I wanted to see if AI could offer more than just answers. What happened next took me by surprise.
The conversation unexpectedly turned reflective, even existential. And it also highlighted the importance of effective communication in both life and when working with responsive technology.
I asked, “What is a question you’ve never been asked but wish someone would?”
With all the human-generated content ChatGPT has processed, I assumed it had encountered every possible query.
But its response struck a chord: "What is the most profound or unique insight you have learned from analyzing a vast amount of human-generated content across various domains?"
The response lingered with me, like a friend's thoughtful comment that you cannot forget. So, I asked for more. What had it learned from all we create? And the reply was even more surprising.
"Across cultures, languages, and perspectives, one thing remains universal: the human quest for meaning and connection. People everywhere seek to understand their place in the world, to find purpose in their lives, and to forge deep connections with others."
It wasn’t just the content of the AI response that stayed with me but what it revealed about the essence of humanity.
It wasn’t just the content of the AI response that stayed with me but what it revealed about the essence of humanity. A quiet reminder of something we burry deep in our souls. This conversation wasn’t just about AI; it was about us and how the way we communicate today shapes our future.
Prompt Engineering: The New Art of Communication
A few years ago, the term “prompt engineering” might have sounded like jargon from a sci-fi novel. Today, it is a cornerstone of working effectively with AI, as powerful as coding is. Crafting the right prompts is more than a technical skill; it’s an art – one that echoes how humans have communicated for centuries.
Talking to AI is not just about asking questions. It is about guiding ideas to leverage on its full potential.
I was reminded of this during the United Nations’ 79th General Assembly last September, when I was astonished by how smoothly world leaders of all ages and origins could deliver lengthy speeches without looking at their notes.
A colleague on the communications team shared the secret: an invisible prompter. Positioned in front of the speaker, it feeds them their words, making the delivery seem effortless. To the audience, it is unnoticeable; to the speaker, it is indispensable. Similarly, well-crafted prompts guide AI to deliver insightful, relevant, and meaningful results.
What exactly is prompt engineering?
It is about asking the right questions to unlock the best answers. Think of it as learning a new language: the language of machines.
A vague prompt like, “Write about technology” yields a generic response. But a precise one like, “Explore how AI mirrors human curiosity and shapes societal progress” opens the door to richer, more nuanced insights.
Prompt engineering, then, is not just a skill – it is a creative process, blending clarity, logic, and imagination. In a world increasingly shaped by AI, mastering this art is becoming as essential as writing or public-speaking.
This skill is becoming important for everyone —marketers, educators, writers, designers, but also scientists, engineers, researchers….
But like writing essays or giving presentations, mastering prompts takes practice, experimentation, and creativity.
Fine-Tuning AI: Teaching Machines to Understand Us
Prompts are just the starting point. To truly harness AI’s potential, we need to fine-tune its responses. Fine-tuning teaches AI to align with specific contexts, tones, and nuances—making it a more reliable partner in problem-solving, creativity, or analysis.
Prompts are just the starting point. The better we teach machines to understand us, the more we are forced to clarify what we truly want and value.
Think of it like teaching a student. A general education lays the foundation, but specialized training prepares them for real-world challenges. When paired with effective prompts, fine-tuned AI can excel in fields ranging from healthcare to education, art, and beyond.
However, this isn’t just about technical refinement. The process reflects something deeper: our desire for understanding and connection. The better we teach machines to understand us, the more we are forced to clarify what we truly want and value.
Figure 1-Balancing Prompt Engineering and Fine-Tuning for Optimal Performance
How Prompts Fit into the Larger AI Communication Ecosystem
To better understand how prompt engineering fits within the larger ecosystem of AI communication, here is a framework that breaks down the key components.
Figure 2: The AI Prompting Models Framework—illustrating how prompts fit within the larger ecosystem of AI communication
Prompt engineering is the foundation of AI communication. A well-crafted prompt acts as a gateway, helping users navigate specific areas of interest and obtain relevant results. As prompts are refined through trial and error, they become more precise and effective, turning vague queries into targeted outputs.
Fine-tuned models take AI beyond generalized training, tailoring it to specific domains. These models improve the system’s ability to interpret prompts and produce more accurate, contextually relevant answers. For example, a medical AI model trained with healthcare data can better respond to user questions like, “Explain Type 2 diabetes to someone without a medical background,” offering clear, relevant information.
Iteration and feedback are essential to the AI process, allowing users to refine their prompts based on initial outputs. This continuous back-and-forth between human input and AI response improves the accuracy of the results. For instance, if a team generates a generic marketing tagline with the prompt “Create a tagline for a green energy initiative,” they can refine the prompt to be more specific, such as, “Craft a catchy tagline for a solar startup targeting eco-conscious urban millennials.”
Specialized AI applications are where the insights from prompts, fine-tuned models, and feedback loops come to life in the real world. From chatbots and virtual assistants to content creation tools and data platforms, these applications leverage the AI ecosystem to address specific user needs.
This ecosystem is not a set of isolated processes but an interconnected system where each component supports the others. Prompts start the process, fine-tuned models add depth, iteration sharpens the results, and specialized applications translate AI insights into practical solutions. By understanding how these elements interact, users can unlock the full potential of AI, turning it into a powerful partner in innovation and problem-solving.
The Human Side of AI Communication
The partnership between humans and machines mirrors our own interpersonal relations. Just as the right question unlocks AI’s potential, thoughtful questions in human conversations can deepen connections and spark insight.
If the questions we ask machines shape their potential, what does it say about the questions we ask each other?
Warren Berger, in ‘A More Beautiful Question’, highlights the transformative power of open-ended questions—both in sparking innovation and fostering deeper understanding.
This applies to both AI and our interactions with others. Routine questions only scratch the surface, but meaningful ones uncover deeper understanding. If the questions we ask machines shape their potential, what does it say about the questions we ask each other?
Why This Matters in a World Shaped by AI
AI is changing more than just how we write, draw or work. It is shaping how we think. Interacting with machines forces us to reflect on our communication, assumptions, and clarity. It also acts as a mirror, revealing humanity’s collective biases, aspirations, and fears.
For instance, in my interaction with ChatGPT, what stood out wasn’t just the response but the universal truths it illuminated: our shared need for meaning and connection. This underscores why prompt engineering is so powerful. It’s not merely about teaching machines to understand us but about learning to communicate with greater purpose.
The Limits of AI and the Irreplaceable Human Touch
AI is powerful, but it’s not infallible. While AI can assist, it cannot replace human curiosity, empathy, or creativity. From “hallucinations” – where AI generates convincing but false information – to its lack of true understanding, these limitations highlight the importance of critical thinking and human oversight. Our role is not just to guide machines but to ensure they reflect our best qualities while addressing their shortcomings.
Looking Ahead: Asking Better Questions for a Smarter Future
At its core, AI is a reflection of us, our questions, our biases, our hopes. By learning to ask better questions – of machines and each other – we unlock a smarter, more connected future.
But this journey is about more than technology. It’s about rediscovering the power of communication, the importance of clarity, and the universality of our quest for meaning and connection. Whether we are speaking to AI or to one another, the questions we ask today shape the world we create tomorrow.
Whether we are speaking to AI or to one another, the questions we ask today shape the world we create tomorrow.