It happened to compilers, operating systems, and renderer. It's gradually happening to game engines.

That's why the largest software companies are really just network effect monopolies. You only use it because other people use it or it relies on some sort of proprietary format.

Many companies like Adobe, Microsoft and Google could sustain a dramatically lower headcount if they started from scratch and never went outside the confines of their network monopoly markets.

From that perspective a one person unicorn is very feasible if we think about it as building a network effect monopoly.

In fact a close example already exists -- Minecraft was very close to being a single person unicorn.

If we have a one person unicorn, I'd imagine it'd be something more like Minecraft than someones newsletter or Google!

Ilya Sutskever's "business" (really his skillset/personal brand) was worth $1B+ the minute he stepped away from OpenAI.

Joe Rogan could fire all his staff and there would still be companies lining up around the block to buy his podcast for $1B+.

There are many such examples of a person having something or being something that is worth a lot to someone else, to the tune of billions.

So by that metric a one-person billion dollar company has already been achieved. What remains are the technicalities (does it need to be incorporated, does it need to have revenue, does it need to operate as a traditional business).

Purely from the operational perspective, you need to handle support, you need to handle billing, to handle lawsuits, to handle infrastructure, to handle incidents... and still have a good idea of what's going on.

So no, not sure it's happening.

I'd be surprised if she felt the same way.

Will there be a shovel that uncovers a billion dollar gold nugget? Why sell the shovel then?

If one person can make a billion dollar company, it likely means that the product itself isn't too complex or big - unless said person has been stealth developing a product for years. But in this case, let us assume it is something founded recently, and aided with AI because, well...AI. That's the pitch right?

So what moat could such a company have? If you can be one man building a billion dollar company in a very short time, hundreds, probably thousands of competitors will spawn in an instant. And everyone will be racing to the bottom to acquire customers.

Ok, so maybe one of those - the most popular, will receive massive VC funding, so that they can simply outspend their competitors into market dominance. But still, that doesn't change the fundamentals: That the product is likely trivial enough to be solo-coded by a bunch of devs out there.

I think it is mostly AI founders hyping up their AI products. "Now you too can become a billionaire overnight, if you just completely integrate these AI tools into your workflow!"

My take is that as the big AI players advance, and the models become better, we'll get some sort of platform where we can just generate whatever tools we need on the spot. That will, to some extent, kill off the LLM wrapper business. In the end, why pay $10 a month for some subscription if you can get the same product directly from the company that makes the machinery. Right now this isn't the case, as they don't have any such tools - but, I'd guess that in a couple of years that is possible.

Near future AI systems, novel research methods and unique datasets (and a lot of luck) could let a single researcher make breakthrough advances. I am thinking of a room-temperature superconductor in materials science or a better transformer architecture in computer science, potentially yielding a billion-dollar patent or company, even if that’s beyond today’s AI system's capabilities.

The specific idea here is not important, the leverage of a new scientific advancement could be though.

Not every company needs to be a SaaS with focus on B2B that won't scale sufficiently for investors with one employee...

To see for yourself how this might be done in principle, ask your favorite LLM to 1. Comprehensively describe the standard operating procedures for a selected corporate role, and to create a comprehensive plan for writing computer code to implement those SOPs.

Here is a link to a ChatGPT 5 query about the role of Chief Customer Success Officer: https://chatgpt.com/share/689b9c31-bb20-8006-89fb-0ad082a52a...

also whatsapp had ~50 employees b4 meta acquired them for billions, imagine what one can do now with aws, ai agents etc erlang/elixir is the only language that made me believe that I could make services that 100s of millions of people could use with no worries and with all the iaas options out there, i wouldnt have to care about managing it and just throw money in the furnace for scale

But you won’t have to be beholden to anyone other than the customer.

Billion dollar valuation depends on investors, investors will never pay that valuation due the inherent risk in a one-person company. They’ll force them to hire for redundancy.

[0] https://en.m.wikipedia.org/wiki/Bus_factor

I say this as someone who spend the last decade building a business with zero employees. I’m nowhere near a billion: by design & the patent absurdity of that “goal”.

But not willing to grow over a 10 person team may be a strong and feasible personal preference.

My bet is on $10 billion with a team of 10 people.

Does that still count?

If we broaden the scope like that, there are many one-person companies doing even more in gaming. Schedule I sold for $100 million in less than six months.

https://www.andycarnevale.com/blog/one-person-billion-dollar...

We already know there are billion dollar humans. Maybe this will be realized by one of those Meta hires, just getting their LLC acqui-hired.

i'm getting the idea though the whoever she is - she is making good amount of money: she should demand a higher bar that I don't think llm's can meet today just because she can afford it and the risk that it matters is too high.

For understanding the guy's point it helps if you know anything about Heather Cox Richardson. Merely saying her copyeditors could be replaced by AI is a gross minimization, probably most of her own work could be too.

HCRs newsletter is popular because she delivers much purer, harder Trump Derangement Syndrome than anyone else on the market. That's it, that's the entire secret to her success. She delivers an ordinary and easily LLMable service (summarizing news), but ramps the Trump hate up orders of magnitude. Her competitive edge is she delivers to a market of liberals who find themselves disgusted with the mainstream journalism because, as they see it, it tries much too hard to be neutral and unbiased. It's sort of like how the Guardian makes millions off of donations by being a very openly left wing paper, except 1000x more extreme.

One might think US journalism is not particularly neutral, but they claim they are, so if you reject the whole notion that journalism should be unbiased as idiotic appeasement the market doesn't have much for you. Into this gap steps Heather Cox Richardson, a humanities academic who for a low low price will send you a daily news report that, every day, tells you how evil and bad Trump is and how good a person you are for realizing that.

Go read some of her reports and then tell us an LLM couldn't write them cold, I dare you. It probably wouldn't even need to be a big LLM. Summarization and rewrites are things they're really good at. She never even cites sources and hallucinations aren't the sort of thing her customer base are going to notice.