AI Is Not a Bicycle. It's Escape Velocity.
"Steve Jobs famously said that computers were bicycles for the brain. AI is a spacecraft travelling at 11.1 km per second for the brain. What happens when humans are not needed to operate the vehicle? Their role may be reduced but the inspiration won't. What inspires a human is a very human thing and will probably never be satiated. Solving problems are not inspiration. They are waypoints on the way to realising inspiration."
The Bicycle Era
In 1980, Jobs watched a study about locomotion efficiency. A condor topped the chart. Humans ranked somewhere in the middle. But a human on a bicycle beat everything. He called the computer a bicycle for the mind.
That metaphor held for forty years. It was accurate. A bicycle amplifies human effort. You still pedal. You still steer. You still decide where to go. The machine multiplies your input. It does not replace it.
Every tool from the spreadsheet to the search engine operated on this principle. Human input in. Amplified output out. The ratio improved. The dependency remained.
The bicycle era is over.
The Spacecraft Era
On May 26, 1969, Apollo 10 re-entered Earth's atmosphere at 11.0794 km per second. That is 39,897 km/h. It remains the fastest speed ever achieved by a crewed vehicle. The crew did not pedal. At that velocity, human reaction time is irrelevant. The physics outrun the biology.
AI operates at cognitive escape velocity. It does not amplify your thinking. It outruns it. A large language model processes, correlates, and generates across more domains, faster, than any human mind. This is not a difference in degree. It is a difference in kind.
A bicycle needs a rider. A spacecraft at 11.1 km per second needs a mission.
The distinction matters. When the tool moves faster than the operator can react, the operator's role changes. They stop steering. They start defining the destination.
When Operators Are Optional
Every technology eventually automates its own operation. The elevator operator disappeared. The switchboard operator disappeared. The typesetter disappeared. Not because humans became unnecessary — because humans became unnecessary at that layer.
AI follows the same trajectory. Today, a skilled operator gets more from an AI system than a novice. That gap closes. It closes fast. The interfaces simplify. The models improve. The prompting becomes invisible. Within years, operating AI requires no more skill than operating a search bar. Within a decade, less.
This is not a threat. It is a phase transition. When operation is automated, the question shifts. It moves from "how do I use this tool?" to "what do I point it at?"
The operator becomes optional. The architect becomes essential.
The Inspiration Bottleneck
Here is where most analysis stops. People see AI automating execution and assume the scarce resource becomes creativity. That is close. It is not precise.
Creativity is the ability to combine existing elements in new ways. AI does this. It does it relentlessly, at scale, without fatigue. Generative models produce novel combinations across text, image, code, music, and video. Creativity, defined as recombination, is no longer scarce.
What AI cannot do is want something.
That distinction — between solving and wanting — is the bottleneck. AI solves. It solves faster and broader than any human. But it solves toward objectives it did not set. Every solution is a waypoint. The destination is chosen by a human who felt something, wanted something, needed something to exist that did not yet exist.
Inspiration is not a computation. It is a drive. It comes from lived experience, from dissatisfaction, from a vision of what should be. No architecture produces this. No training run learns it. It is irreducibly human.
The New Human Role
The shift is structural. For the entire history of technology, humans provided both direction and execution. Sometimes the execution was physical. Sometimes cognitive. But humans did the work.
AI separates these layers cleanly. Execution moves to the machine. Direction stays with the human. Not because we choose to keep it — because the machine cannot take it. You cannot automate wanting.
This changes what matters. Technical skill depreciates. Domain expertise depreciates. The ability to operate tools depreciates. What appreciates is the ability to define missions worth running. To identify problems worth solving. To hold a vision that no dataset contains because it has never existed before.
The human role is not reduced. It is distilled. Everything that could be automated gets automated. What remains is the part that was always the point.
I have spent thirty years building things. Over a hundred patents. Every one started the same way — not with a solution, but with a frustration. Something that should exist and did not. That frustration is the irreducible input. AI handles everything downstream.
The question is not whether AI replaces humans. It does, at the execution layer. The question is whether humans rise to the role that remains. Mission architects. Not bicycle riders. Not spacecraft operators. The ones who decide where the spacecraft goes.
The vehicle is built. The velocity is real. The only scarce thing left is knowing where to point it.