Food for thought... AI bubble?

LINK https://m.youtube.com/watch?v=OuGMNHXONy4&pp=ygURdG9tIGxlZSBmdW5kc3RyYXQ%3D

Tom Lee: Why the Bull Market Isn’t Over Yet | 2025 Market Outlook

Great listen from one of the best. NVDA p/e not even as high as costco or wmt. This is not a bubble

quote:

---

NVDA p/e not even as high as costco or wmt.

---

NVDA PE is 57.2, yes? You didn't read it. No worries.

quote:

---

you first

---

It's a compelling argument he lays out with multiple, too many to read, citations. Seems like the contracts start falling apart when the data centers can't be completed nearly on time or budget. The math ain't mathin' as they say.

But who knows, maybe the innovations that will come from this will accelerate anything this dude is arguing against. Also strikes me that the article ignores anything it's already used in - cloud, API's, software, etc. Certainly there is earnings and/or legitimate savings happening there.

quote:

---

NVDA PE is 57.2, yes? You didn't read it. No worries.

---

On the date of his presentation.. 10/22… costco PE was higher than NVDA PE. No worries bud.

Loading Twitter/X Embed...
If tweet fails to load, click here.

Most definitely a bubble. Mentioned it in a few threads but check out the railroading boom from late 1800s. Similar setup. Critical infrastructure that is needed to advance humanity but there will only be a few winners out of this. The rest will go belly up or get absorbed.

At this point the bubble is so big, the administration may take measures to keep it inflated for the time being. A significant pullback in AI and we are in a recession.

quote:

---

Funny, you just described his point... Only the Mag7 are dropping money into it. There's only about $1B in revenue total outside of them. But anyway...

Sorry I tried to have this conversation :)

---

Oh you wanted a conversation but you just dropped a link with no commentary yourself. Sorry you are awkward in starting conversations.

And how did I describe his point? He literally said "despite the fact there's no real demand for generative AI compute". That is ridiculous.

Then you mention "only Mag7 are dropping money into it" - why do we ignore seven of the biggest companies on earth including the biggest in the industry to conclude "there is no demand"?

Yes I guess if you ignore all the massive companies in the industry that spend billions on generative AI, there is like no demand for it.


And the hyperlink of his ridiculous no real demand statement takes you to another article there he states "Mea Culpa! I have said a few times “$40 billion” is the total amount of AI revenue in 2025, and I need to correct the record. $35 billion is what hyperscalers will make this year (roughly), and when you include OpenAI, Anthropic and other startups, the amount is around $55 billion. If you include neoclouds, this number increases by about $6.1 billion. In any case, this doesn’t dramatically change my thesis. "

Thanks for clarity on $1 billion non mag 7 - and congratulations on Open AI, Antrhopic and other startups (and countless others) making the mag 7 - momentous day!

quote:

---

Absurd statement.

---

It really is ridiculous - as ridiculous as saying there is no demand.

The author is a video game loser that has moved on from writing about video games to trying to convince people work from home was the peak of productivity to now making a name with this.

His statements like "The only thing “powerful” about generative AI is its mythology. The world’s executives, entirely disconnected from labor and actual production, are doing the only thing they know how to — spend a bunch of money and say vague stuff about “AI being the future.” show is is just a whiny little 'whoa the common man' who is bitter he could not move up the ladder.

Other gems "NVIDIA needs this myth to continue, because in truth, all of these data centers are being built for demand that doesn’t exist"

"Because an executive’s job is so vague, they can telegraph the value of their “labor” by spending money on initiatives and making partnerships."

"Generative AI exists for two reasons: to cost money, and to make executives look busy."

Again a thousand pardons for not taking such an insightful author more seriously.

How amazing, AI can help explain:

That’s a great and very relevant question — because the comparison between today’s AI boom and the dot-com bubble (late 1990s–early 2000s) gets thrown around constantly.
But while there are similarities, there are also big structural differences that make today’s AI surge more durable in certain ways — though not immune to corrections.

Here’s a breakdown:

?

?? 1. Revenue Reality vs. Hype

Dot-com bubble:
• Most companies had no real revenue model — just “eyeballs” and website traffic.
• Valuations were based on potential rather than profits.
• Example: Pets.com went public with negative margins and collapsed within months.

AI bubble (today):
• AI leaders like NVIDIA, Microsoft, Amazon, Google, OpenAI, etc. are already generating billions in AI-driven revenue.
• Even smaller AI firms often have real paying customers and cost-cutting benefits from AI integration.
• There’s hype — yes — but also cash flow behind the story.

?

??? 2. Infrastructure Maturity

Dot-com era:
• The internet itself was in its infancy. Broadband, cloud computing, and mobile didn’t exist yet.
• Many ideas (like e-commerce, streaming, online advertising) were too early for the available tech.

AI era:
• The infrastructure already exists: hyperscale data centers, GPUs, cloud platforms, and massive datasets.
• AI is being built on top of a mature internet economy, not a speculative new one.

?

?? 3. Capital Flows & Market Structure

Dot-com bubble:
• Cheap capital + retail speculation (everyone buying IPOs blindly).
• Many IPOs were low-quality startups with no business model.

AI bubble:
• The biggest capital flows are going to established tech giants with diversified income streams (NVIDIA, Microsoft, etc.).
• Venture capital is more cautious than in 1999; fewer pure “AI concept” IPOs exist so far.
• However, some smaller AI startups are starting to attract frothy valuations — that’s the risky frontier.

?

?? 4. Productivity vs. Promise

Dot-com era:
• Internet’s productivity payoff took a decade to show up (after 2005, with cloud, mobile, etc.).

AI era:
• We’re already seeing measurable productivity gains in coding, design, data analysis, logistics, and automation.
• AI tools are being embedded directly into core business workflows, not just speculative consumer ideas.

?

?? 5. Where It Is Similar
• Valuation excess: AI chipmakers and software plays are trading at very high multiples, reminiscent of 1999.
• FOMO investing: Retail and institutional fear of missing out is fueling runaway momentum.
• Winner-take-most dynamics: Just like dot-com had Amazon and eBay emerge from ashes, AI will likely see a few giants survive — but many overhyped players fail.

?

?? 6. Likely Outcome

Instead of a total collapse like 2000, the AI cycle will likely go through:
1. Hype peak (we’re here) ?
2. Mini-crash/correction (wash out weak players) ?
3. Long-term secular growth, as real use cases compound productivity and profits.

Think dot-com 2.0, but with real foundations this time.