The (real) reason why learning AI is hard

The reason "Learning AI is hard" is because it's harder to do something with AI if you can't do it alone.

You can't design with AI if you don't know why designs are good
You can't code with AI if you don't know what code is capable of doing
You can't animate with AI if you don't know why certain animations are good

When all of these things are lacking, it will be harder to use AI. Using AI is limited by your own ability to articulate the output you want. And your ability to articulate an output is limited by the outputs you understand are possible, and your understanding of which outputs are actually good.

1. Why this concept changes AI education

AI is doing something that most educational frameworks haven't caught up to yet: it's stripping away technical execution as a required step in the learning process. You no longer have to grind through the mechanics of a tool to arrive at the outcome. But here's what that doesn't change — it's still just as important to understand what makes a specific end result actually valuable.

If you can understand the details of an end product that make it effective — why a particular animation feels smooth and alive, why a layout draws the eye in a certain direction, why a piece of copy converts — you can communicate that to AI. And that ability to communicate the why behind a great result is, essentially, the real skill now.

When it comes to actually educating yourself on using AI well, there are two main areas of focus:

1. Learning how to speak to AI to create a desired output. This is the craft of prompting — being specific, giving context, describing quality in a way the model can act on.

2. Learning what outputs to actually create. This is the harder one, and it's entirely human. It's taste, judgment, and domain knowledge — understanding what makes a design worth making, or a feature worth building, in the first place.

What this means is that AI education has to be human-first. The technology is not the hard part anymore. The hard part is knowing what to do with it.

2. Combine multiple fields

There is real value in breadth now — more than there ever has been before.

When you had to execute technical tasks manually, depth was everything. You picked a lane — designer, developer, writer, marketer — because mastering even one of those required years of focused effort. Going wide meant going shallow, and shallow work didn't produce good results.

That calculus has changed. Now that you're directing AI rather than executing every task yourself, the more fields you can move through, the more powerful your output becomes. AI has already absorbed a massive amount of the deep technical expertise you used to have to grind out by hand. Which means you can now bring together things that used to require entire teams: software development, brand strategy, visual design, motion graphics, content creation. You can hold the vision across all of it.

If you can build a software product and run the marketing and handle the creative direction and produce the content — and you can do all of that with AI — you can combine it into something that feels like a single, cohesive vision. Because it is. It all comes from one person who understands what they're building and why.

It's actually feasible for a single individual to do that now. That wasn't true five years ago.

3. Most of learning used to be learning the tool

Think about what it actually took to learn graphic design before AI. You couldn't just start making things. First, you had to learn Adobe Illustrator. And learning Illustrator meant learning about vectors, Bézier curves, anchor points, pen tool behavior, straight lines, curved lines, sharp corners, how to shade, how to adjust brightness, how opacity layers interact, how border radii work. A hundred small technical competencies, just to get to the point where you could make a picture that looked intentional.

That's one example. But the pattern holds everywhere. Learning to code meant learning syntax, debugging tools, how to navigate a terminal, how to configure environments. Learning to make music meant learning the DAW, signal chains, compression, EQ, routing. The technical overhead was enormous — and it came before you could do anything creative at all.

Now, if you can accurately describe what you want, you can get the result. The technical layer is not gone — it's just not your job anymore. What's left is the ability to describe quality with enough precision that AI can produce it.

That means a huge portion of what used to be called "education" was actually just tool training. And that portion is now much smaller. What has to fill that space is something more fundamental: understanding what makes a specific thing high quality, and why. That's what AI can't supply on its own — and what you have to bring to the table.

4. Individual agency will soon explode

The ability of the average person to have real power and control over their own life is about to increase by an order of magnitude.

AI enables everyday people — people who have even a modest depth of understanding in any given field — to produce enormous amounts of output in that field. If you understand what good software looks like, you can build software. If you understand what a strong brand looks like, you can build a brand. The gap between having an idea and being able to execute on it has never been smaller.

The implications of this are significant. For owning a business, it means that the barrier to entry for a serious, well-executed operation is now within reach of a single motivated person. For creating art, it means work that used to require a team — and the budget to pay that team — can now span multiple mediums and still feel cohesive because it comes from one vision.

There's a lot of noise online about how the future is bleak. About what AI is going to take away. But I think that framing is getting it backwards. The upside here is genuinely enormous — and it's particularly large for everyday people, not just for big institutions or tech companies.

It's sort of like before airplanes existed. It was genuinely hard to imagine what air travel would look like or what it would mean — not because people were unintelligent, but because nothing in their experience pointed toward it. The reason so many people are afraid of AI right now is probably the same thing: it's harder to imagine the upside when you've never seen it. The downside is easier to picture because it's a variation on things that already exist. The upside is new.

I think the upside is huge. And I think the everyday person who understands how to harness this technology stands to gain more from it than almost anyone. That's what I find genuinely exciting about this moment.

TL;DR