Agriculture is much slower, every iteration may be is a year, or (in tropical climates) half a year. Microelectronics is comparably slow, and even more unforgiving about making mistakes. Building robots does not scale ls easily as producing chips, let alone software.
These areas need a different model of investment, with a longer horizon, slower growth, less influence of fads, better understanding of fundamentals. In some areas, DARPA provided such investment, with a good rate of success.
Leaving behind only those completely focused on returns.
When turning the spotlight to capital that is seeking returns, it is true that these areas may be mediocre places to deploy fresh capital, but it doesn't mean that these players aren't competing, and they will probably be cranking out sheet metal and port cargo logistics optimization well after 90% of the AI startups fold.
The caveat is of course Private Equity, which is about 10 trillion in assets. They can derive high returns from these areas, but it requires leverage.
It is still unsettling seeing Uber turning a profit, but even with that they're not turning a net profit over the lifetime of the company yet and won't be for a few years. Hopefully no-one pops up to compete with them now the sector has profits in it.
What VC capital is not interested in is regular fuel, which can burn steadily and expand gradually, without a shock wave. Such companies can be quite important. Say, GitHub was such a company for many years, before it took a large VC investment and got acquired MS. Investing in such companies requires much more diligence and foresight, maybe too much predictive power to work at mass scale.
VCs' math only works because a single 1000x hit easily pays for a hundred of duds. If ROI per hit is 2-3x, and research required is 10x more deep, the prospects likely start to seem too bleak for folks with billions seeking return.
Just heard of Palmer Luckey. Hmm! Money? No big staff, not much equipment, essentially just one person?? $1B+, quickly? Example: Taylor Swift. Did she ever hear of Linux???
If you have money, the returns you'll get elsewhere are much less attractive, and can only be justified if they're very safe investments.
To fund a similar sized hardware start-up you need a full lab andddd already the proposition is dead.
And engineering teams usually scale up with revenue as well.
I feel like your numbers are the myth that gets told not what actually often happens
Contrast this with biotech venture capital which has been doing well for decades, often investing more capital in a year than software VCs. The difference is that all the research, clinical trial, and manufacturing expertise is already here and concentrated in a few localities like South San Francisco, San Diego, and Boston.
I think that's why most people just aren't that upset about tariffs. It would be nice to be able to participate in our own economy other than by grifting off real estate or software.
I feel like anything relevantly practical is denied investment.
But when it comes to anything flashy and hip, a train of dump trucks filled with cash couldn’t deliver money as quick as the VC dollars that flow into to startups with no business model and no hope of being profitable…
Yeah, I get that startups should invest profits and not actually make profits for a while… But when they’re on their 4th round of funding with thousands of employees… shouldn’t they at least try to be a bit more financially responsible?
They don't know anything else.
https://www.currentmarketvaluation.com/models/s&p500-mean-re...
Exactly. Margins are dropping rapidly: https://ethanding.substack.com/p/openai-burns-the-boats
- The chatbot people have a personal attachment to
- The processing tool.
In the second, you only care about the result. Something like Claude Code can call any other provider if that's cheaper and visa-verse. Once I have the result, my dependency / lock-in is no more than a brand of toilet paper. The providers will have to do the 'capitalism thing' and compete.
It's almost like WeWork's, valuated at IT levels by being in the style, only for investors to eventually figure out the marginal production costs are not reducible to near 0, and you can't just bully out competition / network-effect to get a monopoly.
And this applies to any company that wraps and re-sells AI.
Something the tech-VC world is so unfamiliar with, it's scrambling to present the truth of what is 'econ-101' for the rest of the world.
This YCW18 ag company was acquired less than 3 years in by John Deere for $250M: https://techcrunch.com/2021/08/05/john-deere-buys-autonomous...
That said, allowing VC into 401ks and such I would agree is an abominable idea, because this stuff isn't marked properly until it is in distress. Actually, that area could use better regulation. Volatility laundering is already a systemic risk. Many of these vehicles have creative ways to not mark to their market value, which makes pension fund managers and leered participants happy because it greatly improves the perceived risk metrics and performance, at the equal expense of cloaked fragility.
But perhaps just let them have a thunderdome, and if they want to breach the walls and enter areas like retirement funds where society agrees standard are higher, there is a strong set of filters/regulations that must be adhered to.
