This month I spent $50 on OpenRouter credits. I find it very useful for coding assistance (not coding per se). It's $50 that I never spent before on this kind of service. That's what we call growth - it quite literally came out of nowhere.
This month I reduced my spend on cloud AI and started using local models as primary - it's maybe 2 or 4 times slower for some things but it still is getting the job done just fine. I talked to a friend of mine who's been doing the same running smaller Qwen models locally for coding.
Your $50 isn't enough to sustain trillions of dollars in CapEx investments. This investment cycle - at this scale - only makes sense if within the next 5 years AI replaces something crazy like 5-10% of jobs, if not more. Not changes, replaces completely.
I merely pointed out, that there is real growth behind this - 50$ from my wallet went to this, that did not go towards it at any point in the past. Your argument that my 50$ is not enough is duly noted, but also intellectually lazy.
It's the marketing they were doing all the time that AI will replace developers and other jobs, then the companies started this bullshit firing thousands of people without really thinking through. It's more sane to advertise that AI will augment us and I guess if they followed this path a more sustainable growth is expected to happen!
Their exhuberance at the thought of decimating the labor force overrode their discretion.
Reminds me of this:
>I went through this Ford engine plant about three years ago, when they first opened it.
> There are acres and acres of machines, and here and there you will find a worker standing at a master switchboard, just watching, green and yellow lights blinking off and on, which tell the worker what is happening in the machine.
>One of the management people, with a slightly gleeful tone in his voice said to me, “How are you going to collect union dues from all these machines?”
>And I replied, “You know, that is not what’s bothering me. I’m troubled by the problem of how to sell automobiles to these machines
> I've spent several 1000 dollars on AI products this month, roughly 3000 dollars.
> Trust me, there are many other people like me in the world and the enterprises are spending even more.
Awesome, let's do some math here.
I'll do both $1k/month and $3k/month, please follow along. To make the math even simpler to follow, I'll actually reduce your amounts. ~$800/month will get us about $10k/year and ~1600/month will mean about $20k/year, makes for easier divisions.
From what we know about current AI tech, about ever increasing hardware prices, about ever increasing electricity prices, about the ever increasing DC construction prices, AI companies need to invest a fair chunk of money each year to keep the whole thing going, let's be SUPER conservative and put that amount at $200bn per year.
Then at least another 200 000 000 000 per year = $0.2tn/year.
So $2tn next year divided by $10k/year means that means that they will need 200 million yearly subscriptions to recover the money already invested. At $20 k/year would mean 100 million yearly subscriptions. Spread over 5 years that would mean 40 million yearly subscriptions and 20 million yearly subscriptions.
Then for each year, just to cover the costs, at $10k/year 20 million yearly subscriptions would be needed, and at $20k/year 10 million yearly subscriptions are needed.
So that's 60 million yearly subscriptions and 30 million yearly subscriptions.
I used Claude to extract some numbers. The total global addressable workforce that makes more than $80k per year (where an employer would dare spend $10k/$20k per year on AI) is about 100 million people. The total private population that has $10k/20k per year in disposable income is about 500 million people (excluding China, since they will for sure not use Western AIs en masse).
So that's about 600 million users (just stacking private users + how much companies would pay for their workers).
So just to break even each year, 10% of those would need to pay those crazy high subscriptions, and 5% the extra crazy high subscriptions.
For private users if they get 1% of that rate, it means that a lot of private individuals have fallen on their collective heads.
For enterprises, nobody's going to increase their salary expenses from $80k to $90k-$100k for benefits that we can't even quantity, let alone guarantee an upside of 10-25%. That kind of budget will be allocated for people making $150k or above, which makes the total global addressable workforce something like 50 million, most likely less.
I want to have what you're smoking.
Yes, demand is there. Demand to prop up how much we're investing. NO WAY. At actual prices and actual LLM productivity gains, we should probably be investing 20% of what we're investing.
A lot of people will be wiped because of Nvidia and friends. Even worse, a lot of regular people will suffer because we've distorted our societies so much due to this hype train.
You're on the right track but looking at the numbers incorrectly, specifically, focusing on one example the OP gave. The better approach is to look macroeconomically. Here's the number to look at: Global knowledge worker salaries are at $50 - 70T annually. That is the number enterprises are already paying for knowledge workers.
If AI makes these workers even 1% more productive, that is $500 - 700 billion value annually. At an ongoing annual $0.5T return, a $2T investment (also over the next few years, note) doesn't seem too bad!
Then consider that actual studies from all the way back in 2024, i.e. the era of spicy autocomplete, before agents landed on the scene, put the productivity boosts much higher, like 30% or more. (Interestingly, this is corroborated by survey based data from the St. Lous Fed: https://www.genaiadoptiontracker.com/) Even assuming a conservative average boost of 10%, that is $5 - 7T value annually.
Add how many ever grains of salt you want to those numbers, the investment is nowhere near as out of whack to the potential revenues as people fear. This is why everybody from Big Tech to VCs to entire nation states are desperately scrambling to get in on the action.
1. There's no way China will let any of the Western frontier labs in, so that's probably 1/3 out of those $50-70tn that they'll never touch.
2. It turns out that LLMs are more of a commodity than expected because the basic tech is basically "Attention is all you need" plus a few things everyone has access to (mixture of experts, caching, batching, etc). So yes, it's a "winner take most" market, but there will likely be a healthy base of cheap models so the "collection" (price gouging) part of the cycle (or enshittification) will be hard to execute.
3. Either hardware remains expensive, in which case every N years entire DCs have to be rebuilt and then Capex needs to flood in - think highway systems being rebuilt, but instead of every 20-30 years for highways, here we'd be talking every 5-7 years.
4. Or hardware becomes cheap in which case cheap LLM hosters are competitive and problem #2 is even worse. Or the nightmare scenario for all of these investment scenarios, local LLMs become viable for most people.
Even if China accounts for a whopping 50% of the knowledge work market, the TAM is still $25 - 35 trillions. The US alone is $10 - 11T. Any fraction of that is still a huge number that can recoup $2T in a few years.
I agree LLMs are already a commodity market, definitely at the non-frontier model level, but I don't think it affects monetization prospects much. After all, server compute is a commodity and yet cloud businesses have been exploding even before AI.
And compute is exactly why it won't be a "winner takes most" market. It is clear now that compute capacity is and will likely remain the biggest moat. Looking at Claude Code is instructive; arguably it was the better product, but it kept going down so much that Codex and other competitors have gained on it. Similarly, China could have the best models, but its access to hardware is deliberately limited by geopolitics, so it's likely their threat will be manageable for a while yet.
A key part of the success of AI companies will be in securing hardware and operating that infra cost-effectively via economies of scale. Hardware will remain expensive for a long time yet because all the hyper-scalers and neo-clouds are severely crunched, and all the fabs (mostly TSMC) are already at capacity even as demand keeps exploding.
And for better or worse, most of that supply will still flow through Nvidia, despite attempts from competitors like TPUs and NPUs, for the simple reason that Nvidia has the monopoly profits to outbid everyone else on the real chokepoint, which is fab capacity.
Oh, that's easy. Bernie will pick 11 volunteers and together they will be dropped off behind enemy lines. Let's call them "the Dirty Dozen". They will capture the evil bastards who violated American law outside of US borders (that's just evil) and bring them to Justice™ (Impossible is nothing. Call 1-800-JUSTICE. Conditions apply).
> the obvious international treaty
That's even easier. We just take the international treaty on cloud computing as a base and...
The point is an arms race is bad for everyone. "I don't know what WW3 will be fought with, but I do know WW4 will be fought with sticks and stones" type of stuff.
The Cold War was an arms race that gave us wonderful things: amazing technologies we still use today (GPS, the Internet) and the collapse of the soviet union.
"Were we on the brink of that thing that we already did in Japan and nothing happened a few times? Oh my god, I need to hide in a cave and never do anything outside again"
Of course not. Even the developers of AI admit there's a huge risk though, no? Even they are calling for a slowdown. Do you think you know more than they do?
I know well enough; I'd be surprised if they know more than I do. And I also know very heavily that we cannot stall into societies of empowered deficient humans.
Can you name a "Marxist-Leninist AI" that's made by a real AI lab (i.e. no finetunes of open models made by someone on the internet)? I'm just trying to understand what the other side's equivalent of MechaHitler is here.
Can you name a "Nazi AI" that's made by a real AI lab (i.e. no finetunes of open models made by someone on the internet)? I'm just trying to understand what the other side's equivalent of MechaStalin is here.
Can I remind you that the MechaHitler episode was a real thing? If an LLM being lobotomized to the point of supporting Hitler out of nowhere wasn't Nazist in your opinion, then nothing is.
Tay is not an LLM and was trained directly by its user base. Grok is trained by xAI that obviously wanted to turn it into a far-right talking point reciter but went just a little too far.
What? I started the conversation, I set its boundaries. You're the one who's now trying to redefine it. I challenged the parent to show me a Marxist-Leninist AI, trained by an AI lab in a way that's equivalent to what xAI did.
I don't think you even have a point. The AfD doing the same thing doesn't say anything, and I never said there can only be one pro-Nazi AI in the world. I just dismissed Tay out of hand because 4chan users spamming a primitive chatbot with neonazi shit has very little relevance to the conversation about AI labs trying to influence their products to support extreme ideologies.
That would require one to have robots, leaving Americans and Chinese in the mix. Unless you mean biorobots, then it's nothing new and there's no need to be alarmed.
> For all the feigned outrage, people want this, and will work hard every day to pay for the next iteration of the surveillance apparatus.
My sad realization is that the Buddhists were right all along. The primary drivers of human mind, before self-purification, are greed, anger, and ignorance. What you are describing here is greed. People want to get useful things for free. This is the core of the problem. If people were, by default, expecting to pay for everything they use, spying on them would not be accepted. Instead, they want stuff for free and they choose to be willfully blind to the spying.
> If people were, by default, expecting to pay for everything they use, spying on them would not be accepted.
The spying may have started with free products, but it has metastasized into every aspect of life, free or not. It always does.
> choose to be willfully blind to the spying
Many don't. They also, by and large, don't have a choice. Data brokers, Flock cameras, WiFi-based radar[1]. You really can only do so much as an individual before the only escape is being a hermit in a cave.
> The primary drivers of human mind, before self-purification, are greed, anger, and ignorance
Which then gets distilled into the only drivers of corporations. Corporations that then destroy all moral options for individuals.
Funnily enough I've found for the most part the counter to all this spying also tends to be free products. Signal, pi-hole, ublock origin, graphene OS, Opnsense, Linux, etc. I have yet to find paid-for apps that are superior to the free ones in this regard. If I want an image editor I'm using Gimp, not Photoshop. Libreoffice is better than Microsoft Office if you don't want to be spied on. Most of the cost is learning about the alternatives and how to use them.
You can't fight with the times. Kids want colorful mods, dance moves, and other stuff like that. Girls also play, they won't play things like the original Doom or Quake, and boys don't get to meet girls much if they are not playing the same games.
reply