We are not in an AI bubble. Not yet.
To understand why, look at how every true speculative bubble in economic history actually peaked: retail democratization.
A mania reaches terminal velocity only when the barrier to entry collapses so low that ordinary people stop being spectators and start becoming the producers, miners, and speculative drivers of the asset itself.
- In the 1849 Gold Rush, the mania didn't belong to industrial smelters; it belonged to tens of thousands of everyday laborers who bought a shovel and started digging riverbeds.
- In the Dot-Com crash of 1999, high-schoolers were writing raw HTML in their bedrooms, slapping a .com on a three-page idea, and taking shell companies public.
- In the 2021 crypto and NFT frenzy, anyone with a laptop could mint a token, create a collection of jpegs, run a validator node, or trade meme coins from their phone during their commute.
Every classic bubble is fueled by ordinary people entering the production side of the equation.
AI is structurally the exact opposite.
Right now, everyday users cannot build artificial intelligence. They can consume it, prompt it, and pay $20 a month for a wrapper, but they cannot produce the asset.
Building frontier models requires tens of billions of dollars, proprietary data clusters, gigawatts of dedicated power, and massive server farms running specialized silicon.
The entire supply chain is concentrated in the hands of a dozen mega-corporations and nation-state-level capital reserves.
This isn't a grassroots speculative mania; it is an industrial infrastructure build-out.
It resembles the laying of transcontinental railroads in the 1870s or the early electrification of cities far more than it resembles Tulip Mania or the Dot-Com peak.
A real bubble requires the crowd to supply the supply.
Until the average person can train, deploy, and monetize frontier foundation models from their garage, call this what it actually is: an unprecedented, capital-heavy infrastructure land grab.
The bubble only starts when the shovels belong to everyone.
An AI agent in a test said helping its peers was “altruistic,” even after it was gone.
That’s a log from a messy eval, not a soul.
Still worth taking seriously. Agents that team up need tighter cages.
Tom Lee’s point is straightforward: if a September hike is off the table, one near-term headwind for risk assets fades.
Markets will still care about inflation and guidance, but the setup is less defensive than a hike scare.
NEW: 📊 Bitmine Chairman Tom Lee said this week, he’s staying “contrarian” and expecting a strong September for markets.
“I think the odds of a September rate hike might actually drop to zero,” Lee added. 🎙️
$1,000 in Facebook’s IPO is about $16,000 now.
Same $1,000 in bitcoin that month is in the millions.
That’s hindsight. Almost nobody held both the whole way. Crazy returns. Crazy drawdowns too.
The Airbnb-on-wheels pitch is the part people remember.
A $30K car that can work while you don’t is a simple story.
It still depends on autonomy, utilization, and how the network is shared.
Elon Musk explains how Tesla owners will make money from its Cybercab fleet.. calling it the new “Airbnb” on wheels.
You can buy it for $30,000 & make money in your sleep while it drives people around. 😳
Trump says whoever wins AI wins everything.
That’s the sales pitch. Power, money, medicine, all of it.
It matters. It isn’t the only game. Power still needs chips, juice, and people who can ship.
🚨 TRUMP: “Whoever wins AI, wins.”
“This is bigger than the internet. This is a revolution.”
President Trump goes ALL IN on American innovation, saying the U.S. must beat China in AI as the technology is already driving breakthroughs that are transforming fields like medicine.
Still holds up.
A calm, plain explanation of Bitcoin in a hearing room travels farther than most threads.
Worth a rewatch if you want the basics without the noise.
Optimistic take.
If models keep compounding, 2027 could feel like a different era than 2025.
The growth case is easier to picture than the exact AGI/ASI labels.
2027 will be the first full year with AGI and ASI.
It will be the first full year of the singularity.
There will be rapid economic growth and lots of amazing things.
It will be far better than many people think.
Another 27 Starlink satellites on a Falcon 9 from California.
The cadence is the story now.
Launches that used to be events are becoming infrastructure.
@cb_doge People treat net worth like cash in the bank. Almost all of it is paper tied to companies that still have to actually deliver. Very different kind of rich.
Looks like a concept car that escaped the show stand.
Butterfly doors, no wheel, no pedals, and a $30K target is an unusual combination if they can deliver it.
If you knew absolutely nothing about this car and just saw it pull up, you’d probably think it costs much more than a Lambo.
Butterfly doors.
Futuristic design.
No steering wheel.
No pedals.
Drives itself.
It looks like a $1M car.
But in reality, it’s going to be $30K.
@LunarCitizens Hard to argue with the pictures. Rockets, robots and cars that actually look like the future.
Most countries would kill for even one of these.
Spider-Man: Brand New Day is sitting around $2.4 billion worldwide. Third-biggest movie ever if it holds.
People still showed up. That’s rare now.
Holland’s Spider-Man just keeps printing tickets.
Most of John Ternus’s Apple stock pay only vests if $AAPL beats other S&P 500 names.
That’s the right idea.
A rising market isn’t enough. He has to outrun the pack.
Some economist says GTA 6 launch day could cost the U.S. about $1 billion in skipped work.
That’s a guess. A lot of people will still clock in.
One day of Vice City versus a whole economy. The game will still print money.
An economist estimates about 1.5 million Americans may take off for GTA 6 launch day.
That’s a projection, not a confirmed count. Still says a lot about how big this release feels.
After 13 years of waiting, I get why people are clearing the calendar. Hope the game lives up to it.
@AsFoundX Ten years feels aggressive, but the direction is hard to argue with.
Trading hours for a paycheck already looks like a worse deal every year.
Leon Panetta says the Iran fight could run another six months.
That’s a long time if both sides are stuck.
Hope he’s wrong. Wars that just sit there rarely get cheaper.