AI doesn't make you smarter. It adjusts the rate at which your thinking can become a thing in the world. That's a different claim, and the distinction matters.
There's a gap between the speed of thought and the speed of making. You can think of ten ideas before you've written down the first one. You can imagine a building in less time than it takes to describe a single room. The mind moves at whatever speed it moves. The body, and the work of making things, moves considerably slower.
This is what changed for me when I started using AI: not what I think, but how fast I can make what I think real.
I want to be specific about this, because the popular story about AI is often about intelligence: AI as smarter-than-you, as a replacement for thinking, as something you defer to. That's not my experience and I don't think it's accurate. What AI actually does, at least for creative and technical work, is adjust the rate. It's a rate adjustment, not an intelligence amplifier.
A bionic limb doesn't make you better at deciding what to pick up. It closes the gap between the decision and the act.
A useful comparison: a bionic limb. A prosthetic arm doesn't make you better at deciding what to pick up. It doesn't improve your judgment about what's worth reaching for. What it does is close the gap between the decision and the act, between intending to reach and the reaching actually happening. The intelligence was never the bottleneck. The execution was.
For a long time, execution was the thing that stopped most ideas from becoming real. Not lack of capability. Not lack of ideas. The bottleneck was the actual making: the typing, the coding, the building, the iteration, the ten hours of technical work between "I want to try this" and "this exists." Most ideas died there. Not because they weren't worth pursuing. Because the cost of pursuing them was prohibitive.
AI changes that ratio.
I should be clear about what I mean when I say AI helps. Taste is still mine. Judgment is still mine. Direction is still mine. AI can't want something. It doesn't know which of my ten ideas is the one worth pursuing. It doesn't feel when something is almost right but not quite, when a sentence is a little too clean, when a structure works technically but misses something. Those things are not the bottleneck. They're the point.
What AI does is let me test ten ideas in the time it used to take to test one. Which means I actually find out whether they're good. Because I actually finish them.
The Fractal Story Engine is a good example of this. It's a nonlinear fiction archive: nine territories of speculative fiction, each built from a single premise that fractures into multiple forms. The architecture was always clear to me: the structure, the voice, the visual system, the way premises generate worlds. What AI changed was the rate at which that architecture could become actual pages. Hundreds of specific, particular pieces of prose, each one its own thing. The writing is mine: the ideas, the characters, the sentences, the decisions about what each story is doing. The execution gap closed.
That's the thing I keep coming back to. It's not about being smarter. It's about finding out, in your actual lifetime, whether the ideas you have are any good. The execution gap used to mean that most ideas never got tested. The ones that survived were the ones you cared about enough to fight through the friction. That's a meaningful filter, and I'm not saying friction was worthless. But a lot of good ideas died in that gap too.
When the gap closes, something different becomes possible. Not just more output. A different relationship to the ideas themselves. You can follow the tenth idea instead of stopping at the second because you ran out of steam. You can find out what you actually think, because you can actually finish the thought.
Minds faster than bodies. That was always the condition. It just became, for the first time, something you could do something about.