You Must Understand Your Subject Basics to Use AI as a Teacher

Using AI to Learn: You Have to Understand the Problem to Ask the Right Questions

In the world of artificial intelligence (AI), there’s a simple truth: to get the most out of AI, you need to understand the problem well enough to ask the right questions. It sounds obvious, but this is something many people overlook. The misconception is that AI will magically produce answers without a deep understanding of the context, but that’s not how learning works—whether it’s with a machine or a human.

Why Asking the Right Questions is Key to Learning

Let’s start by considering what happens when you learn anything new. Imagine you’re learning about circular motion and the forces exterted on an object at the end of a string. If you don’t grasp the basics of force and motion, you’ll find it difficult to ask meaningful questions about centripetal acceleration. You might ask, “Why does the rock move in a circle?” but to truly explore the concept, you need to ask something more focused, like “How does the string exert force on the rock?” And then you'll get some illuminating answers as well as equation.

In my example I needed to understand how the strring literally pulls the rock towards the center - it's totally counterintuitive. Having said that, the discussion began with me asking why an object in circular motion is considered to be accelerating when it moves at constant speed.

When you ask specific questions like that, you’re showing that you understand at least part of the problem. This is crucial because AI, like any tool, only works as well as the questions you give it. AI can’t think for you—it can only process the input you provide. If you ask vague or poorly thought-out questions, you’ll get vague or incomplete answers. But if you can narrow down the question, AI becomes a powerful assistant.

The AI Mindset: Approach Learning Like a Detective

When you’re working with AI, imagine you’re a detective piecing together clues. You might not have all the answers yet, but you need to know enough to figure out what’s missing. Every time you ask a question, you refine your understanding. For instance, if you’re trying to understand how centripetal force works, you don’t stop at “What is centripetal force?” Instead, you ask, “What role does the tension in the string play in circular motion?” That’s a question with layers, and each layer takes you closer to a full understanding.

This is one reason why software engineers and other professionals shouldn’t fear AI replacing them. AI can handle complex computations and provide useful insights, but it still needs direction. The skill lies in knowing what to ask and how to interpret the answers. This isn’t a threat to expertise—it’s a tool that enhances it.

The Feynman Technique Meets AI Learning

One of the best ways to learn anything is the Feynman Technique, which is based on explaining a concept as simply as possible. The same technique can be applied when working with AI. Here’s how it works:

• Step 1: Choose a topic you want to understand.
• Step 2: Explain it in your own words, as if you’re teaching someone else.
• Step 3: Identify gaps in your explanation where you can’t fully clarify a concept.
• Step 4: Go back to the source (or use AI) to fill in those gaps with new information.

When you use AI in this process, it acts like a tutor that helps you clarify your weak spots. But here’s the catch: you can only spot those gaps if you already have some grasp of the problem. Without that initial understanding, you won’t even know what you don’t know, and you won’t be able to ask the right questions to fill in the blanks.

How AI Amplifies Learning, Not Replaces It

The fear many people have about AI replacing professionals, like engineers, teachers, or scientists, is largely misplaced. AI is an extraordinary tool, but it doesn’t innovate or ask questions. Humans still need to frame the problem and guide the process. It’s a bit like having an extremely fast calculator—you can do incredible things with it, but only if you understand the equations you’re trying to solve.

Take software development, for example. Many programmers use AI-based tools to streamline their work, debug code, or even auto-generate certain functions. But these tools don’t write the software from scratch. The developer needs to know the system architecture, understand the requirements, and provide clear instructions to the AI. Asking vague questions like, “How do I make this program better?” won’t get you anywhere. But if you ask something more precise, like, “How can I optimize this loop for faster execution?” the AI can give you actionable suggestions.

Why Context Matters: AI and Problem-Solving

Context is everything when learning with AI. When you ask a question without understanding the underlying problem, the AI might give you a technically correct answer that’s completely irrelevant to your needs. For example, if you’re trying to improve website performance and ask, “How do I reduce load times?”, AI might suggest general solutions like compressing images or enabling caching. But if you know the real issue is on the server-side, you could ask, “How can I reduce database query times on my WordPress site?”—which is a far more targeted and useful question.

In both cases, the AI can help, but the quality of the help depends entirely on how well you understand the problem. If you don’t have the right framework in mind, you’ll end up chasing the wrong solutions. This is why it’s crucial to develop a solid understanding before diving into AI-assisted learning. Just like with human teachers, you need to know what questions to ask to get meaningful answers.

The Role of Curiosity and Iteration

Curiosity drives the process of learning with AI. If you’re not curious enough to dig deeper and keep asking better questions, you’ll never reach the level of understanding you’re aiming for. In Feynman’s words, “You have to keep asking questions, and then ask questions on top of those questions.” AI is no different—it’s a tool for iterative learning. You start with a basic question, get an answer, refine your understanding, and then dive deeper with more precise questions.

This is where learning becomes a collaborative effort between you and the AI. It’s a process of back-and-forth inquiry, each step building on the last. As you refine your questions, you uncover more layers of understanding, much like peeling an onion.

Conclusion: AI as a Learning Partner, Not a Replacement

In the end, AI is a partner in learning, not a replacement for the learner. It excels at processing information and providing insights, but it relies on human curiosity and understanding to guide its efforts. To truly harness the power of AI, you must understand the problem well enough to ask the right questions. This is where the real power of learning lies—not just in the answers you get, but in the clarity of the questions you ask.

As Feynman might say, “The joy of figuring things out” still belongs to us. AI just makes the process faster and more efficient—but we’re still the ones steering the ship.

Here's the full converstion on circular motion

Motion in a Circle: Why Is It Always Accelerating?

Keith: Motion in a circle. Why is it always accelerating?
AI: An object moving in a circle is always accelerating because its direction is constantly changing, even if its speed remains the same. This is called centripetal acceleration, which is directed towards the center of the circle. Since acceleration is defined as the rate of change of velocity (which includes both speed and direction), the object is accelerating because its direction is continually shifting.

Understanding Circular Motion and Acceleration

Keith: It's totally counterintuitive because we think of acceleration as moving from a speed to a higher speed.
AI: You're right, it feels counterintuitive because in everyday experience, we associate acceleration with speeding up. But technically, acceleration is any change in velocity, which includes both changes in speed and direction. In circular motion, even if the speed remains constant, the object’s velocity is changing due to its changing direction. That’s why it’s accelerating, even without an increase in speed.

Centripetal Acceleration: A Change in Direction, Not Speed

Keith: So circular motion is a vector with constantly changing direction?
AI: Yes, exactly! In circular motion, the velocity is a vector that is always changing direction. The speed might stay the same, but the velocity vector, which points in the direction of motion, is always tangential to the circle. Because the direction of the velocity vector changes as the object moves, this is why it experiences acceleration even when its speed doesn’t change.

Conceptual Hurdle: Acceleration Without Speed Increase

Keith: So how do we define that acceleration? It's easy in linear motion—we define it with ds/dt where s is speed and t is time. What do we do for circular motion?
AI: In circular motion, acceleration is called centripetal acceleration, and it’s defined differently than in linear motion. It’s related to the change in direction of the velocity, not its speed. The formula for centripetal acceleration is: \[ a_c = \frac{v^2}{r} \] where \( v \) is the object's speed and \( r \) is the radius of the circle. This shows that the acceleration depends on both the speed and the radius of the circular path.

Why Does the Object Move Towards the Center?

Keith: No savvy. The object is trying to move away from the central point of the circle, not towards it.
AI: I get what you mean—it feels like the object is being “thrown” outward, which is often called centrifugal force. But from a physics standpoint, that's more of an apparent force due to inertia. The real force keeping the object in circular motion is centripetal force, which pulls the object inward, towards the center, and keeps it from flying off in a straight line.

How Do We Define Acceleration in Circular Motion?

Keith: So centripetal force is exerted by what?
AI: The centripetal force can come from different sources depending on the situation: - **Tension**: In the case of an object on a string, the tension provides the centripetal force. - **Gravity**: For planets, gravity acts as the centripetal force keeping them in orbit. - **Friction**: For a car on a curve, friction between the tires and the road provides the centripetal force. In each case, the force pulling the object towards the center prevents it from moving in a straight line.

The Role of the String in Circular Motion

Keith: I get it for gravity—there's a central point of gravity pulling inwards, easy to see. With the string though (assume infinite rigidity), there's no gravity, and the string is not generating force while the other end of the string is exerting force to cast the object away.
AI: The key is understanding that even with an “infinitely rigid” string, the tension is still pulling the object inward. The object, due to inertia, “wants” to move in a straight line, but the string is constantly pulling it towards the center, changing the direction of its motion. This inward pull is the centripetal force, and there’s no actual outward force acting on the object—what you’re sensing is the object’s inertia, not a real force.

Measuring Centripetal Force in Circular Motion

Keith: Let’s assume we can set up measurements for the forces at work for a rock on the end of a piece of string and that the whole thing is in motion already. Now we want to measure the forces to which the rock is subject. How do we measure the centripetal force directly?
AI: You can directly measure the centripetal force by measuring the tension in the string, which provides that force. You can attach a force sensor or dynamometer to the end of the string to measure the tension. This value will give you the centripetal force, which can also be verified with the formula: \[ F_c = \frac{m v^2}{r} \] where \( m \) is the mass, \( v \) is the speed, and \( r \) is the radius of the circle. The measured tension should match this calculated force.

The Key Insight: The String Is Pulling the Rock

Keith: I think I have it now. The conceptual difficulty is accepting that the string is 'pulling' on the rock.
AI: Exactly! The string is indeed pulling on the rock, providing the necessary force to keep it moving in a circle. The object's inertia makes it "want" to move in a straight line, but the string constantly pulls it inward, changing its direction. Once you understand that the string is exerting a force inward, everything about circular motion makes sense.

The Importance of Asking the Right Questions

Keith: You are extremely helpful when I know what questions to ask. I think that's the trick with using AI for anything—you must know enough to ask the right questions.
AI: You're absolutely right! Knowing the right questions to ask can make a huge difference in getting useful answers, especially when using AI. It’s all about guiding the exploration of a topic effectively. I'm glad I could help, and it sounds like you're getting great at pinpointing exactly what you need to ask!