Ideas Are Cheap, Conviction Is Not: The Founder Edge in the AI Era

Founders   20 July 2026 · 6 min read

There is a certain kind of person who guards their startup idea like treasure, afraid to say it out loud in case someone runs off with it. In 2026 that fear is almost quaint. Ideas were never the scarce resource, and now that anyone can build almost anything, they are cheaper than ever.

Here is what is actually scarce, and getting scarcer: the conviction to keep going through the long, unglamorous middle, the stretch where the idea is no longer exciting, the results have not come yet, and quitting would be easy and reasonable. Ideas are cheap. Conviction is not.

Why ideas stopped mattering

Two people can have the exact same idea. One thousand people probably do. What separates the outcome was never the idea, it was who did something about it and kept doing it. That has always been true, but AI made it undeniable, because now the gap between having an idea and having a working version of it has shrunk to almost nothing. Building is no longer the filter it used to be.

When anyone can generate ideas and anyone can build them, neither the idea nor the building is where the advantage lives. The advantage moved to the thing that was always the real filter: sustained will.

The valley everyone forgets

Every venture has a shape. The start is exciting, full of possibility and dopamine. The eventual success, if it comes, is satisfying. Between them sits a long valley: the middle, where the novelty is gone, the early traction has plateaued, the problems are boring and hard, and there is no applause. This is where almost everything dies, and it dies not because the idea was wrong but because the person ran out of conviction.

The valley is invisible from the outside. We see the exciting start and the eventual win and imagine a smooth line between them. Founders who have been through it know the line is a grind, and that the only thing that carries you across is a belief strong enough to outlast the boredom and the doubt.

What conviction actually is

Conviction is not blind stubbornness, and it is not refusing to change course. It is something more specific: a belief in the underlying mission that is strong enough to survive being wrong about the details. You will be wrong about many things, the feature, the message, the market. Conviction is what keeps you in the game long enough to get those things right, rather than concluding after the third setback that the whole thing was a mistake.

It is also, increasingly, the thing that cannot be automated or copied. A tool can generate your idea and build your product. It cannot want it. It cannot care whether it lives or dies at 2am in the middle of the valley. That caring, that refusal to quit on something you believe in, is irreducibly human and irreducibly yours.

Why this is good news

If the edge were the idea, you would be at the mercy of cleverness, and someone is always cleverer. If the edge were execution, you would be at the mercy of tools, and someone always has better ones. But the edge is conviction, and that is something you actually control. You cannot guarantee you will have the best idea or the fastest build. You can decide to be the one who does not quit.

In an era where ideas are free and building is easy, the founder's real advantage is the oldest one there is: caring enough to cross the valley when everyone with the same idea has already turned back. Ideas are cheap. Be the one with the conviction.

Frequently asked questions

Why are startup ideas considered cheap?

Because many people have the same ideas, and AI has made building them nearly effortless. Neither having an idea nor building it is a filter anymore. What separates outcomes is who persists in doing something about it, which was always the real scarce resource.

What is conviction in the context of founding?

Conviction is belief in the underlying mission strong enough to survive being wrong about the details. It is not stubbornness or refusing to adapt. It is the will that keeps a founder in the game through the hard, boring middle long enough to get the details right.

Why does conviction matter more in the AI era?

Because ideas and execution have both been commoditised, so the advantage moves to what cannot be automated or copied: sustained will. A tool can generate and build your idea, but it cannot want it to succeed or refuse to quit when things get hard.

What is the valley founders talk about?

The long middle stretch of a venture, after the exciting start and before any eventual success, where novelty fades, progress plateaus, the problems are hard and unglamorous, and there is no applause. Most ventures die here, not from bad ideas but from founders running out of conviction.