Ha! That's an amusing interpretation of how supply and demand interact to find an equilibrium price/quantity point.

A typical Econ 101 textbook says:

    higher demand --> higher price
    higher price --> higher supply
    higher supply --> lower price
    lower price --> higher demand

In Econ 101 we pretend that cycle eventually reaches an equilibrium. In grad school we analyze the dynamics.

But even if we believed your demand causes supply causes zero price model, why is that unique to engineering and not business acumen? And why do programmers get paid anything, don't they work for free?

Yes and if you've actually paid attention you will see that a high salary signals other workers to flood the market with a lower salary thus starting the race to the bottom :)

Programmers have largely driven themselves to zero, just take a look at the workers on freelance websites and how much difficulty a native English speaking freelancer has against an army of commoditized labor.

There are still six digit earning engineers and always will be but you can't look at that as a metric for the massive commoditization that has occurred with generic programming in the past 15 years.

There's some premium for face-to-face, but it turns out that even with remote work, price is still a good indication of quality.

I'm a freelancer and fully booked, so maybe those low cost competitors aren't so competitive after all.

Another nonsense implication of your theory: it's better to have a low wage, so you stay under the radar and don't signal other workers to enter the market.

For the foreseeable future, machine learning experts will be in sufficient supply for them (i) to have wages which are considerably above zero and (ii) have the ability to raise capital from people that don't think they're commodity labour

(it cuts the other way too: if an oversupply of labour in a field as difficult-to-learn as ML does arise, the demand shortage causing wages to fall is almost certainly because ML techniques aren't giving companies and traders as much of a financial edge as they hoped for...)

Added bonus for the ML experts: in a lot of the possible areas they can choose to work in, their individual marginal contribution to the company's profitability is at least as quantifiable as that of a salesperson. In many areas of finance you can quantify the impact of an individual line of code!

It's not enough to hire a machine learning expert if you are planning on building out your own capable ML operation confident in competing against the best and the brightest at Azure, Google & AWS. You'd need a team of ML experts.

On the other hand, it's easier to piggy back off the hard work of highly paid experts in this space now commoditized into 1/10th of a cent per application of an ML algorithm from Microsoft or Google or IBM.

Commoditization is going to be in full force as businesses realize they can just have a junior developer to plugin IBM Watson or Azure than hire an expensive PhD student who may know everything under the sun but cannot compete against an army of highly paid experts completely under the control of a highly collusive labor market.

I think that idea is largely responsible for the decline of HP, IBM, etc.

Karpathy was offered in excess of a million out of school, I'm going to assume that you don't know what you're talking about.

Silly google, they must've skipped Econ 101 since they hire tons of engineers and are doing TERRIBLY!

And how many Karpathy is there working for non-Google? Mariah Carrey made $140 million dollar record deal with Sony Music, there's boat load of money to make in the music scene!

So your SOLE contribution is capital, right?

Why would I go to you and not a VC firm? I sure as hell don't want to work for someone like you....

(oh and there are plenty of others like Karpathy, like Facebook, Baidu, Microsoft, Uber, Amazon, a metric fuckton of startups, etc. who are being paid ridiculous amounts of money to build nets)

I think you are just trolling at this point.