ProblemExplorer
ProblemExplorer on Hacker News
The primary difference is that starting a company means talking to people.
So while you can be inside a room, read books and try things out at a computer and become a great engineer, to start a company, you have to go out and talk to people - but the steps are exactly the same. The feedback loop is slower and far more noisy.
Computers don't lie but people do, and that's the essence of business.
What would be the collateral a lender(s) would use to loan NVIDIA money in 1993 when it had no assets, nor revenue, nor a proven business plan?
Even if a lender(s) provided a loan without any collateral, what would be the terms of that loan? What would be the amount of loan, who would it be underwritten against, what would be the APR?
The overall perception I see most popular on HN of "AI" is that it's a waste of money, natural resources and we should be "doing something else".
I'm old enough where I've personally seen multiple AI winters and also been through major disruptions - the PC era, the consumer internet era, the cloud, social media, gig economy and now "AI".
For every major disruption, the biggest naysayers have been the very groups that got disrupted. They eventually had to either retire or find something else to do.
There's no logical explanation or data you can provide to these naysayers that will change their mind. They have dug their teeth into a narrative disconnected from actual evidence and that's now their identity.
As a result, I have a strong hypothesis that AI will forever redefine what knowledge work - including software - looks like and entails. Credentialism and rentseeking by educational institutions will be a thing of the past. Human experts will be largely employed by frontier and specialized AI agent companies and labs.
I have been extremely lucky to have invested early in Cisco, Google, Amazon, Dell, Uber and NVIDIA - wasn't very sure when I made those investments so much so that I invested only funds I could afford to loose.
I'm now fully leveraging my investments into AI. IF AI goes to h*ll, the consumer, retail, commodities, energy markets are so strongly coupled to it, that I doubt anyone will come out of it for less than a 20% correction, and I will be way past 100 when it does recover, so I have nothing to lose.
I disagree. Google is doing a phenomenal job at capturing use cases at the edge, right where "normie" consumers live. Google's models absolutely suck at coding but if you haven't tried Google Transcribe, you should.
Google has also, far larger distribution and stickiness than Meta.
it's less about getting sued and more about not being blocked from being able to scrape.
> More importantly, how does my vague notion of the ideas are good or bad tie back to my original point that investors willing to loan instead of taking ownership is perhaps more sane?
First, no one is forcing you at gunpoint to respond to me, so feel free to engage and go as deep as you want, as much as you care to, or not at all.
Second, investing, typically in shares of a company, is very different from giving a loan to the same company. Loans are typically based off collaterals, even if it's "unsecured" - it's always secured by something, even if as nebulous as just a credit history or "ability to pay".
No one reasonable in 2026 will argue collaterals with sama or OpenAI today, but I assure you, there was an entire universe of companies that would not give sama a loan in the early 2000s when he struggled to live paycheck to paycheck, or even OpenAI itself (a loan) in 2016 when they were exploring household and gym robots. I'll make the exact same argument for NVIDIA late 2018 to early 2020 post the cryptocurrency demand correction or even before that 2010 to 2015 - go look at the tickers back then.
So again, with the complete power of retrospect:
- Is NVIDIA a company you would loan your own money to?
- Is Dropbox a company you would loan your own money to?
If you get caught up in details like "how long is the term" or "what's the expected ROI of the loan", be generous with those details.
> It’s capital intensive, doesn’t easily scale without linear costs, is basically a commodity, seemingly has negative cashflow forever, and customers aren’t willing to pay what it actually costs. It’s basically the worst tech business model and yet somehow VCs caught a spell and threw every available penny into the fire.
We already agree on the first two. So the following remains:
"commodity, and customers aren’t willing to pay what it actually costs"
To that I say:
- AWS, Uber are all commodity
- Most customers of AWS, Uber aren’t willing to pay what AWS, Uber charges them
Disagreements?
Both can be true! Assume that it costs you $10 to support inference of 1MM tokens, and you sell that for $100. So you have a 90% profit margin.
Now you also have flat rate subscriptions for $200/month, where you sell upto 100MM tokens. That costs you $1000, but you're selling it as a loss leader because that drives up demand, bottom up - for your product, which is good for long-term profitability specially when your investors are not yet sensitive to you burning cash.
I have multiple $500/mo OAI subs. Still cheaper than buying my own hardware for equivalent performance.
This is overly dramatic and pointing that out is necessary because this could cause someone else, even if not you, tremendous anxiety and pessimism about the future: None of which is good for your health.
First of all, no one's been forcing anyone to use AI or "move everything to the cloud". AI is tremendous leverage. It's exactly as useful as a jet or FSD is for commuting. It expedites significantly intelligence-related tasks. No one is forcing anyone to use AI. However, just like a person who owns a jet gets to a place faster than a person without, or a FSD allows you to get work done instead of staring at other cars, a person using AI will get their tasks done quicker than a person without. But that doesn't force a necessity.
Moving "everything" to the cloud is also not going to be financially smart. Companies like Apple, Google, Cloudflare, etc are aggressively moving compute to the edge, locally on device. This increases your own chances of running services locally on device because the device manufacturers will build their hardware to match the demand.
Second, compute capacity has always been elastic. Supply constraints have forever been an issue and that's how markets work. Supply gets constrained, suppliers react to demand, market reaches equilibrium.
In 1996, I had a real tough time getting a T1 line to my office. By 1998, it was no longer an issue. By 2000 there was more fiber than we knew what to do with.
I had a real tough time hiring good, quality engineers 2020 through 2024. That was a constraint in the supply. The market reacted and now there are a lot more good, quality engineers.
Yes, there's a severe constraint now but the market is going to respond and we will have a very different situation in 2030.
Capitalism works. It's not always roses though and not everyone benefits all the time.
Is the idea of a company or companies that R&D and sell GPUs good?
Is the idea of Dropbox good?
Is the idea of diffusion LLMs good?
I am unaware of who Klippenstein is and whether he's left or right leaning. This article via HN is my first exposure to him, but without any context at all, just by reading the article, I can figure out he's very left leaning.
Why does that matter?
It matters because journalism in 2026 has turned from reporting facts into appealing to a certain spectrum of an audience. It's no longer about providing facts of the matter but about leading an audience to a conclusion that they already had in their minds and re-enforcing it. This is the same echo chamber loop journalists accuse social media of but they are guilty of the very same.
---
First, Klippenstein (seems to) imply that it was the Trump administration that began claims of Russian intelligence interference.
Not so. In October 7, 2016 - the Obama administration, via both Department of Homeland Security and Office of the Director of National Intelligence, publicly attributed compromises of political organizations’ emails to the Russian government, connecting the subsequent disclosures to Russian-directed efforts. This was a government intelligence attribution and not merely a statement from the Democratic Party or Clinton campaign
Next, Klippenstein (seems to) imply that CYBERCOM is running the show here. That's not how INT works - CYBERCOM complements, and doesn't replaces FBI/DHS, while CIA is itself foreign focused. CYBERCOM’s role also includes disrupting foreign cyber operations, not just merely investigating them.
Now a very noble interpretation of Klippenstein is he's describing a bipartisan willingness to restrict speech in the name of election security - that his central objection is to the government’s intervention in political expression, not exclusively to which party controls it.
This is a fair argument but if there's foreign intelligence interference, that brainwashes the American public, making them draw conclusions that are factually invalid, whose responsibility is to atleast point that out?
The DoW has, at its disposal, incredibly advanced intelligence dissemination tools (as a requirement for INT operations) that are a great fit for this usecase. Should not the American public, who pay for all of it, not make use of it?
Every scammer, fraudster, scammer-in-training, fraudster-in-training, uses a Google account.
The first company federal investigators often go for data is often Google because that's the start of their lead into their investigation. Google needs to provide answers, sometimes by regulation in countries.
I would say you are either very young or unfamiliar with Uber as a business. Uber was and continues to be extremely capital intensive and you need to qualify your "while". Uber burned cash for 14 years.
The one argument I can agree with is Uber didn't burn as much cash as Anthropic or OpenAI does but they are very different businesses with very different impact on humanity.
yes and that's the entire point of Silicon Valley.
It's also why you can't have another Silicon Valley in Europe or Asia where almost no one is willing to suspend belief and reality like this.
I've been invited to meetings and conferences from places in Sweden, Amsterdam to Thailand, and investors in all of those places balk at the kind of mindset you need to invest in startups. They are more comfortable with loans than investing. They lose confidence faster or require a higher ROI than would make sense. SV is really special and the reason why the best talent from all the over the world gather there.
VCs become rich not because they are smarter and better than Warren Buffet but because they are willing to place bets that Buffet would never take on.
Known to someone who doesn't trade on a heavily AI indexed market (in 2026 all companies, even Tyson or Chevron is an AI play but not as heavily AI indexed as those say 2 degrees away from Anthropic or OpenAI) - likely not.
So if the point you're making is your typical barber doesn't know about this, I am guilty.
If you want to do your own research into this, there are a lot of private reports and publications on the topic.
If you don't want to spend any money, there is even a recent public analysis using a 92% API gross margin.
Anthropic's October 5, 2026 analysis also points to a 92% API gross margin that SemiAnalysis based their report on.
OpenAI reported paid-user compute margin reached 70% in October 2025 - a year ago.
DeepSeek has similar reports. The frontier labs are comfortably in the fat inference margin territory.
Do you have information to the contrary?
Elsewhere on this page I argued how this reminds me of the PC wars of the mid 90s.
We are all overreacting over a nascent industry, upset that it's increasing prices for everyone and pushing humans out of jobs. This is understandable but the conclusion is wrong. We are letting emotions drive decisions.
Depends on how old you are. This analogy makes sense if you were a teenager in the mid 90s. I am older than that so to me, this same pattern had happened when PCs were becoming mainstream. That was 40 years ago so memory of it might be faint or lost. Also VCs weren't really a huge deal back then either but the fundamentals were the same.
We are all overreacting over a nascent industry, upset that it's increasing prices for everyone and pushing humans out of jobs. This is understandable but the conclusion is wrong. We are letting emotions drive decisions.
Uber and Lyft literally "sell rides for a fraction of what all this costs them" upto early 2025. I don't know how old you are but I started to use them back in 2012. I was one of their early users when you could sit next to the driver. From then upto 2025, Uber and Lyft used to literally cost me less than what just gas would cost me to drive the same distance.
This is true too: "It would then claim it’s worth 100x the valuation of the taxi service" - there was literally a website (might be up even now) that literally made this argument and why Uber was doomed to fail anytime now.
Your first two points are also true for Uber Lease - a product that was rolled out for drivers - they could choose to lease specific models and Uber in turn signed agreements with specific automakers. An exact parallel for these points, esp. the circular financing is hard to make because they are not even the same product but I imagine one would agree, on the whole, we are in the conception stage of LLMs and this industry is brand new and figuring things out.
This same pattern had happened when PCs were becoming mainstream. That was 40 years ago so memory of it might be faint or lost. Also VCs weren't really a huge deal back then either but the fundamentals were the same.
We are all overreacting over a nascent industry, upset that it's increasing prices for everyone and pushing humans out of jobs. This is understandable but the conclusion is wrong. We are letting emotions drive decisions.
> It’s capital intensive, doesn’t easily scale without linear costs, is basically a commodity, and customers aren’t willing to pay what it actually costs
What are some VC funded companies that come to mind that do not fit this profile?
A better report would have been how the "Economics of AI Wrappers Don't Work".
Character.ai was likely paying frontier LLM API rates (which are known to be ridiculously expensive at 90% margins) and then had their own R&D plus lawsuit overhead. Most people can use a $20 Claude or ChatGPT sub to build a completely custom Character.ai. Once Character.ai employed MBA Strategy #3 (aka enshittification), churn killed the company (as it should)
---
What doesn't work is attempting to build an AI Wrapper. Those companies had a good run 2022-2025, but with how good both open weights and cheap frontier models have become, AI Wrappers can no longer demand fat margins.
I'm sure people will now argue what exactly AI Wrappers are in the comments below, but for now I'm more interested in understanding what the HN hivemind's take is on the topic as a whole.
I don't use windows but have to run windows 11 VM to access special devices. I don't want to bother with licenses, although I have them, and creating a new incus QEMU VM from scratch everytime from an ISO takes ~5m, freeing me up from worrying about all the activation and license drama.
If the activation and license setup was easy/automated in QEMU, I would imagine snapshots would be way easier. I would prefer not needing to connect to the internet.
Touche. Guilty as charged. I have been very frustrated with a lot of writing that seems to put forth "AI" as something all new on its own.
I cannot imagine why. To me, it's just cheaper labor and all the baggage that comes with that.
I have been hiring and working with outsourced talent for decades and the progression of "AI" has been eerily the same. Around 2000, quality of outsourced talent was poor with a few exceptions. They had a hard time producing a good product unless micromanaged. Their error rates were frustratingly high and it required a lot of late nights (for me) and back and forth.
Outsourced talent from the exact same cities and countries are now a night and day difference. It's like talking to a co-worker - the work quality is now on par but the price has gone up significantly too.
This transformation took about 26 years.
This is why to me, "AI" is just cheaper labor and all the baggage that comes with that. It's following the same trajectory, on an expedited basis.
Thoughts?
I'm responding to your comment because I need your help in rephrasing my comment above to make my point more clear, because the way I wrote it obviously misled you.
No, I didn't say the extreme (what would the extreme even be?).
What I literally wrote down is what I said, which is that using anything (be it a tool that is not autonomous or a fully autonomous agent like a very smart, capable human being) requires skill.
Imagine you and I are working on a team project together. In majority of real-life situations I'm not just going to point at a writing with my fingers to you and from there you figure out what needs to get done and you would magically do it without any further feedback. That's not how anything remotely useful and productive in this world works.
What's more likely to happen is we have a conversation, figure out how to break the task down, together, how to coordinate, how to cooperate, and how to stitch our results back together to complete the job.
As you can see this requires skill at every point.
Hence, the point I was making is that the article is either clickbait or the author doesn't understand how the world works: anything remotely useful and productive in this world requires skill (be it the use of a dumb tool like a welding torch or cooperation, working with something as smart as a very capable autonomous agent or human being).
what does an 100% automated economy look like?
"The transmission in your car: Using It Well Takes Skill."
Further, Claiming AI is a tool is overly reductive. It's as much a tool as a full self driving car is a car.
You can treat the full self driving car, as a car, driving it with full attention and effort like you would have a Camry but then you paid for features you're not using. Also, some full self driving cars might not even have the steering and braking controls as a Camry. That's not the point of a full self driving car and this warped perspective will make you lose out to people who use a full self driving car well, to its utmost, safe capability.
---
So what is the "AI" we are so excited about currently?
That "AI" is most likely an intelligent (semi-autonomous?) agent.
Depending on the intelligence, it can be as useful as a dog, scanning your camera feed for threats and identifying real threats for you, separating activity from tree leaves moving due to the wind or a moving shadow cast by a flying bird.
Or it can be your co-founder, interviewing potential customers, finding and exploring their pain-points, helping you figure out if there's a gap in the market you can build a business around.
I would go even far as to say, AI is having such a moment because AI are cheap intelligent agents - anything else that's remotely close are likely humans that cost atleast a few dollars an hour.
All of these require skill on your part - there's nothing special or unique about AI that makes skill special just like operating the transmission in your car requires skill.
---
In the age of ~AI~ SI, "expensive" is a state of mind.