AI Founder Weekly

levelsio says the riches in the niches playbook is dead

He argues AI companies now fill every niche and the general case at once, leaving most SaaS with nowhere to hide.

levelsio has written off the oldest piece of bootstrapper advice there is. In a reply to Klaas (@forgebitz), who posted that everyone is building everything and that it will destroy most SaaS companies from the inside, levelsio argued that the riches in the niches strategy no longer works, because AI is different from every technological shift that came before it.

@levelsio
@levelsio
X
Everyone's been checkmated by big tech and big AI
Nov 17, 2026 · View on X
Klaas
@forgebitz
X
everyone is building everything this will destroy most saas companies from the inside
Nov 17, 2026 · View on X

His description of the old world is the one most founders learned. You picked a tiny slice, you specialised, and you could be confident no big tech company would dedicate the resources to come after it. His example is a $300M niche where taking 1% gets you $3M a year. Going general was the mistake, because that is where big companies had the budget and the people, and spreading yourself thin got you death by a thousand paper cuts.

What changed, in his telling, is that the big AI companies are not shipping a SaaS product at all. They are selling general intelligence, so they cover any niche and the general case simultaneously without consciously deciding to, and every model upgrade widens the coverage for free.

The trap he describes

That leaves two doors and he thinks both are shut. Stay in the niche, which is what the economics used to say you should do, and the models absorb it. Go general and offer everything, which works for a while until customers work out they can go straight to whoever makes the model the SaaS is reselling, often cheaper or free.

He calls the result the creative destruction of the century, and allows that there will be survivors and new companies, though he says nobody knows yet what they will look like. He then undercuts his own optimism in the same post, saying that part is wishful thinking to avoid getting mobbed, and he is not even sure about it.

Our read, this is an argument, not evidence. levelsio is not pointing at a SaaS that died this way or a churn number, he is reasoning from where model capability is heading. The obvious counter is that distribution, trust and being the place a customer already has their data are not things a model vendor inherits by getting smarter. It also sits next to his earlier point that building has never been easier while making money has never been harder, which is the same worry from the supply side.

Steal the numbers, not the vibes

One email a day. Under three minutes. Real MRR, real pricing, real launches from founders building with AI.

Get the next one by email

Founders building with AI, every day. Real revenue, real pricing, real launches and what they would redo. No hype sludge. Every number links out.