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.
PE is arguably much worse than VC. VC's business model is well understood and by taking VC funding, you are committing to their expected returns.
PE is, usually, unsolicited and is designed to exploit what appears to be a "lazy" balance sheet, but which is actually a stable business producing output and providing a reasonable ROI.
PE has very few redeeming features.
In the US there have been a few, i.e., apparently less than 20, universities with an applied math program up to date in and teaching optimization.
Sooooo, anyone at all seriously interested, long, for decades, would, could, should visit some of those math programs, meet some of the profs, get recommendations for their former students, call them, chat, and offer a job better than their current lawn mowing, fast food restaurant kitchen cleaning, or car washing. Instead of just the US, might also consider Waterloo in Canada. Actually the Chair of my Ph.D. orals committee specialized in optimization in logistics. After sending 1000+ beautifully written resume copies and hearing back nothing, can begin to conclude that optimization is not a hot field and for highly dedicated optimizers who want to sleep on a cot in a single room, forgo bathing, most days eat bread, other days peanut butter, have no children, wife, or family contact, don't own a car, and must get any needed medical care from some of the last resort special clinics. Ah, real optimization!
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.
I especially dislike the way VC funded startups use VC dollars to effectively be a “loss leaders” for years to choke out the rest of the market.
Who wants to risk their own capital or privately pooled funds against THAT?
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.
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.