Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

The AI bubble argument has the causality backwards. Cerebras CEO Andrew Feldman was in the 1990s fiber optic wave. He knows what a bubble looks like. In fiber, supply was built ahead of demand - companies bet on a future that hadn't arrived. AI infrastructure is the opposite: data... show more

Podcast Alpha

6,219 subscribers

16,010 views • 2 months ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Bubbles happen when supply gets ahead of demand. Cerebras CEO Andrew Feldman says AI is the opposite. Everyone is chasing demand that already exists, with constraints in every direction. Bubbles don't usually start with a shortage.

Bubbles happen when supply gets ahead of demand. Cerebras CEO Andrew Feldman says AI is the opposite. Everyone is chasing demand that already exists, with constraints in every direction. Bubbles don't usually start with a shortage.

This Week in AI

11,500 views • 1 month ago

During the dotcom bubble, 97% of America’s fiber was dark because infrastructure was built years ahead of demand. Today’s AI buildout looks nothing like that. The comment “there are no dark GPUs” means every GPU produced is already reserved or monetized the moment it ships. This is the difference between speculative infrastructure & fully utilized infrastructure.

During the dotcom bubble, 97% of America’s fiber was dark because infrastructure was built years ahead of demand. Today’s AI buildout looks nothing like that. The comment “there are no dark GPUs” means every GPU produced is already reserved or monetized the moment it ships. This is the difference between speculative infrastructure & fully utilized infrastructure.

Shay Boloor

545,174 views • 9 months ago

Gavin Baker built a massive position in $ALAB around the AI infrastructure theme, and the stock has now crossed $300. He thinks AI is different and that we are NOT in a bubble. He expresses this thesis behind many of his top holdings, including $NVDA, $MU, and $AMZN.

Gavin Baker built a massive position in $ALAB around the AI infrastructure theme, and the stock has now crossed $300. He thinks AI is different and that we are NOT in a bubble. He expresses this thesis behind many of his top holdings, including $NVDA, $MU, and $AMZN.

Sam Badawi

450,333 views • 3 months ago

Jeff Bezos just made the most counterintuitive argument in tech. An AI crash wouldn’t destroy the future. It would fund it. Jeff Bezos: “If we go back 25 years ago when the internet was in that bubble-ish moment, no one would have predicted a lot of the industrial benefits.” The dot-com bubble erased trillions in market value. Companies that raised hundreds of millions were gone within months. The money vanished. The infrastructure didn’t. Bezos: “All of that fiber optic cable that got laid, and by the way, the companies who laid all that cable went out of business.” Billions worth of fiber optic cable buried under oceans and across continents. Laid by companies that no longer exist. They went bankrupt. The cable stayed in the ground. Bezos: “Like literally went bankrupt. But the fiber optic cable was still there. And we got to use it.” Amazon. Google. Netflix. Uber. Every cloud platform. Every streaming service. All built on infrastructure paid for by dead companies. They funded the future. They just didn’t survive long enough to see it. That exact pattern is about to repeat. Hundreds of billions are flooding into AI infrastructure right now. Data centers. Chip fabrication. Power generation. Cooling systems. Some of the companies writing those checks will not exist in five years. The market will correct. Valuations will crater. The bubble narrative will be everywhere. And the infrastructure will still be standing. Data centers don’t vanish when the stock price hits zero. GPUs don’t disappear when the company folds. Power grids don’t downgrade when investors pull out. Every dollar being spent right now is permanently reshaping the physical world. It doesn’t matter which companies survive to use it. The bubble isn’t the risk. The bubble is the funding mechanism. The railroad bubble overbuilt track that connected a continent. The telegraph bubble laid wire that enabled global communication. The dot-com bubble buried fiber that carries the modern internet. Each time, the investors lost. Each time, civilization gained. AI is that same pattern running at a scale we’ve never seen. The crash will feel like a catastrophe. In hindsight, it will look like something else entirely. The largest involuntary infrastructure investment in human history. The companies that fail will have already served their purpose. The compute layer stays. And the survivors build on top of it. The question was never whether the AI bubble will pop. It’s who will be standing in the rubble with a blueprint.

Jeff Bezos just made the most counterintuitive argument in tech. An AI crash wouldn’t destroy the future. It would fund it. Jeff Bezos: “If we go back 25 years ago when the internet was in that bubble-ish moment, no one would have predicted a lot of the industrial benefits.” The dot-com bubble erased trillions in market value. Companies that raised hundreds of millions were gone within months. The money vanished. The infrastructure didn’t. Bezos: “All of that fiber optic cable that got laid, and by the way, the companies who laid all that cable went out of business.” Billions worth of fiber optic cable buried under oceans and across continents. Laid by companies that no longer exist. They went bankrupt. The cable stayed in the ground. Bezos: “Like literally went bankrupt. But the fiber optic cable was still there. And we got to use it.” Amazon. Google. Netflix. Uber. Every cloud platform. Every streaming service. All built on infrastructure paid for by dead companies. They funded the future. They just didn’t survive long enough to see it. That exact pattern is about to repeat. Hundreds of billions are flooding into AI infrastructure right now. Data centers. Chip fabrication. Power generation. Cooling systems. Some of the companies writing those checks will not exist in five years. The market will correct. Valuations will crater. The bubble narrative will be everywhere. And the infrastructure will still be standing. Data centers don’t vanish when the stock price hits zero. GPUs don’t disappear when the company folds. Power grids don’t downgrade when investors pull out. Every dollar being spent right now is permanently reshaping the physical world. It doesn’t matter which companies survive to use it. The bubble isn’t the risk. The bubble is the funding mechanism. The railroad bubble overbuilt track that connected a continent. The telegraph bubble laid wire that enabled global communication. The dot-com bubble buried fiber that carries the modern internet. Each time, the investors lost. Each time, civilization gained. AI is that same pattern running at a scale we’ve never seen. The crash will feel like a catastrophe. In hindsight, it will look like something else entirely. The largest involuntary infrastructure investment in human history. The companies that fail will have already served their purpose. The compute layer stays. And the survivors build on top of it. The question was never whether the AI bubble will pop. It’s who will be standing in the rubble with a blueprint.

Dustin

140,512 views • 4 months ago

Aswath Damodaran on AI bubble: “Dot-com was an equity bubble. AI is becoming an infrastructure debt bubble.” In the 1990s, many internet companies raised money, built websites, burned cash, and disappeared. When the bust came, shareholders got wiped out. AI is different, because this boom is not just apps and pitch decks. It is data centers, GPUs, power contracts, fiber, leases, project finance, and private credit. The numbers are wild: ↳ Goldman Sachs estimates hyperscalers may spend ~$5.3T on AI/data center capex from 2025–2030 ↳ Morgan Stanley estimates ~$2.9T in global data center construction from 2025–2028 ↳ Top hyperscalers are projected to spend ~$600B in 2026 alone, with roughly 75% tied to AI infrastructure Dot-com asked investors to believe in traffic. AI is now asking lenders to believe in utilization. If demand keeps compounding, the buildout looks genius. But if demand disappoints, the problem is not just “stocks go down.” It becomes stranded capacity. Refinancing pressure. Private credit stress. Defaults. AI may end up becoming the defining technology platform of our lifetime. But the bubble risk is not where most people are looking.

Aswath Damodaran on AI bubble: “Dot-com was an equity bubble. AI is becoming an infrastructure debt bubble.” In the 1990s, many internet companies raised money, built websites, burned cash, and disappeared. When the bust came, shareholders got wiped out. AI is different, because this boom is not just apps and pitch decks. It is data centers, GPUs, power contracts, fiber, leases, project finance, and private credit. The numbers are wild: ↳ Goldman Sachs estimates hyperscalers may spend ~$5.3T on AI/data center capex from 2025–2030 ↳ Morgan Stanley estimates ~$2.9T in global data center construction from 2025–2028 ↳ Top hyperscalers are projected to spend ~$600B in 2026 alone, with roughly 75% tied to AI infrastructure Dot-com asked investors to believe in traffic. AI is now asking lenders to believe in utilization. If demand keeps compounding, the buildout looks genius. But if demand disappoints, the problem is not just “stocks go down.” It becomes stranded capacity. Refinancing pressure. Private credit stress. Defaults. AI may end up becoming the defining technology platform of our lifetime. But the bubble risk is not where most people are looking.

Linas Beliūnas

41,948 views • 2 months ago

Brad Gerstner on 1999/2025 “dark fiber” bubble comparisons: “There are no dark GPUs.” “Everybody's looking for the bubble.” “We've got this pattern recognition, everybody's like, ‘It's dark fiber. It's gotta be dark fiber. There's no way that you can 10x your revenues at Nvidia without it being a bubble.’ But the reality today is Nvidia trades at a lower p/e multiple than it did when it was $200/share.” “Think about what the term dark fiber means. We were laying fiber in the ground that remained dark. We did not have the demand for the fiber.” “There is not a dark GPU in the world today. There's not going to be a dark GPU in the world next year.” “The tokens are being consumed. Every consumer in the world wants answers, they don't want to use 10 blue links.” “Every enterprise wants to deploy this to make their enterprise better.” “National sovereigns want to do this in order to advance their military and other things.” “There's no dark GPUs in the world today.” Brad Gerstner

Brad Gerstner on 1999/2025 “dark fiber” bubble comparisons: “There are no dark GPUs.” “Everybody's looking for the bubble.” “We've got this pattern recognition, everybody's like, ‘It's dark fiber. It's gotta be dark fiber. There's no way that you can 10x your revenues at Nvidia without it being a bubble.’ But the reality today is Nvidia trades at a lower p/e multiple than it did when it was $200/share.” “Think about what the term dark fiber means. We were laying fiber in the ground that remained dark. We did not have the demand for the fiber.” “There is not a dark GPU in the world today. There's not going to be a dark GPU in the world next year.” “The tokens are being consumed. Every consumer in the world wants answers, they don't want to use 10 blue links.” “Every enterprise wants to deploy this to make their enterprise better.” “National sovereigns want to do this in order to advance their military and other things.” “There's no dark GPUs in the world today.” Brad Gerstner

The All-In Podcast

174,963 views • 10 months ago

How do you stop a fiber-optic drone that is immune to EW? The answer is kinetic — you have to shoot it down. But hitting such a small, high-speed target has been the primary bottleneck for defenders. A Brave1 participant has developed a solution: a compact, AI-powered turret that detects, tracks, and neutralizes drones autonomously. The operator simply authorizes the strike. Of course, the turret is not limited to countering fiber-optic drones — it effectively downs FPVs, Molniyas and other UAV that the enemy deploys against our forces. Watch the first footage of the system in action on the battlefield.

How do you stop a fiber-optic drone that is immune to EW? The answer is kinetic — you have to shoot it down. But hitting such a small, high-speed target has been the primary bottleneck for defenders. A Brave1 participant has developed a solution: a compact, AI-powered turret that detects, tracks, and neutralizes drones autonomously. The operator simply authorizes the strike. Of course, the turret is not limited to countering fiber-optic drones — it effectively downs FPVs, Molniyas and other UAV that the enemy deploys against our forces. Watch the first footage of the system in action on the battlefield.

BRAVE1

73,581 views • 3 months ago

Why it is BS that we are in an AI infra bubble... "We can't build data centers fast enough to keep up with demand. We're not building ahead of demand. We're building behind demand. We are trying to keep up with demand, not the other way around." Andrew Feldman What does no one know that everyone should know when it comes to AI infra spending? Brad Gerstner Gavin Baker Nathan Benaich mmurph Alexander Embiricos

Why it is BS that we are in an AI infra bubble... "We can't build data centers fast enough to keep up with demand. We're not building ahead of demand. We're building behind demand. We are trying to keep up with demand, not the other way around." Andrew Feldman What does no one know that everyone should know when it comes to AI infra spending? Brad Gerstner Gavin Baker Nathan Benaich mmurph Alexander Embiricos

Harry Stebbings

10,163 views • 2 months ago

I think the $ORCL earnings has two truths at the same time. The first truth is that Oracle’s AI demand is real, massive and still accelerating but the second truth is that Oracle’s AI demand is expensive to deliver which meansthe stock now comes down to capex, financing, margins and dilution. The opportunity is massive because with ~$638B of backlog, Oracle now has the largest contracted backlog in tech and Larry Ellison is clearly betting the company’s future on OCI.

I think the $ORCL earnings has two truths at the same time. The first truth is that Oracle’s AI demand is real, massive and still accelerating but the second truth is that Oracle’s AI demand is expensive to deliver which meansthe stock now comes down to capex, financing, margins and dilution. The opportunity is massive because with ~$638B of backlog, Oracle now has the largest contracted backlog in tech and Larry Ellison is clearly betting the company’s future on OCI.

Shay Boloor

134,760 views • 2 months ago

We need a bubble so there is enough capital to build out the AI infrastructure needed for the next few decades. It just doesn't seem like we are in a bubble yet.

We need a bubble so there is enough capital to build out the AI infrastructure needed for the next few decades. It just doesn't seem like we are in a bubble yet.

Anthony Pompliano 🌪

876,770 views • 3 months ago

Everyone's asking if we're in an AI bubble. Gavin Baker gave a reasoned take at a16z Runtime on why this isn't like the 2000s. "At the peak of the [Internet] bubble, 97% of the fiber that had been laid in America was dark. Contrast that with today - there are no dark GPUs."

Everyone's asking if we're in an AI bubble. Gavin Baker gave a reasoned take at a16z Runtime on why this isn't like the 2000s. "At the peak of the [Internet] bubble, 97% of the fiber that had been laid in America was dark. Contrast that with today - there are no dark GPUs."

Justine Moore

195,907 views • 9 months ago

Greg Brockman, President of OpenAI, said there is not enough compute in the world to satisfy AI demand, and OpenAI itself cannot launch products it has already built because it cannot find the infrastructure to run them (Save this). OpenAI is spending $50 billion on compute in 2026 alone and it still is not enough. That is the setup but here is the trade. Nebius is one of the most asymmetric infrastructure plays in public markets right now, and most people have never heard of it. Q1 2026 revenue came in at $399 million, up 684% year over year, with AI cloud revenue specifically growing 841% in a single quarter. The company entered 2026 with an exit ARR of $1.25 billion and is targeting $7 to $9 billion by year end, a number that would make it one of the fastest revenue ramps in the history of public infrastructure companies. The contracted backlog sits at $50 billion anchored by a $17.4 billion agreement with Microsoft through 2031 and a $27 billion five-year deal with Meta. They are decade-scale infrastructure commitments from the two largest enterprise AI spenders on earth, signed before the demand curve has even reached its steepest point. Nvidia took a direct equity stake in Nebius, one of only two neoclouds it has invested in alongside CoreWeave. That relationship is not just financial but rather means Nebius gets preferential access to GPU allocation at a moment when every lab and every hyperscaler is competing for the same constrained supply. Contracted power capacity now exceeds 3.5 gigawatts, with expansion plans targeting 5 to 6 GW by mid-2029. And power is the other binding constraint in AI infrastructure, you cannot build a data center without it and Nebius has already secured the capacity that competitors are still fighting to acquire. At full ramp, analysts project revenue in the $15 to $25 billion range by 2029, against a current market cap the contracted backlog alone already dwarfs. Come join Milk Road Pro and get our full Nebius deep-dive, the exact price levels we are watching, how we are sizing the position against the backlog and power capacity timeline, and our full AI thesis. link below!

Greg Brockman, President of OpenAI, said there is not enough compute in the world to satisfy AI demand, and OpenAI itself cannot launch products it has already built because it cannot find the infrastructure to run them (Save this). OpenAI is spending $50 billion on compute in 2026 alone and it still is not enough. That is the setup but here is the trade. Nebius is one of the most asymmetric infrastructure plays in public markets right now, and most people have never heard of it. Q1 2026 revenue came in at $399 million, up 684% year over year, with AI cloud revenue specifically growing 841% in a single quarter. The company entered 2026 with an exit ARR of $1.25 billion and is targeting $7 to $9 billion by year end, a number that would make it one of the fastest revenue ramps in the history of public infrastructure companies. The contracted backlog sits at $50 billion anchored by a $17.4 billion agreement with Microsoft through 2031 and a $27 billion five-year deal with Meta. They are decade-scale infrastructure commitments from the two largest enterprise AI spenders on earth, signed before the demand curve has even reached its steepest point. Nvidia took a direct equity stake in Nebius, one of only two neoclouds it has invested in alongside CoreWeave. That relationship is not just financial but rather means Nebius gets preferential access to GPU allocation at a moment when every lab and every hyperscaler is competing for the same constrained supply. Contracted power capacity now exceeds 3.5 gigawatts, with expansion plans targeting 5 to 6 GW by mid-2029. And power is the other binding constraint in AI infrastructure, you cannot build a data center without it and Nebius has already secured the capacity that competitors are still fighting to acquire. At full ramp, analysts project revenue in the $15 to $25 billion range by 2029, against a current market cap the contracted backlog alone already dwarfs. Come join Milk Road Pro and get our full Nebius deep-dive, the exact price levels we are watching, how we are sizing the position against the backlog and power capacity timeline, and our full AI thesis. link below!

Milk Road AI

14,578 views • 2 months ago

Sundar Pichai, the CEO of Alphabet, was asked point blank if AI is a bubble. He didn't dodge. He said yes -- partly -- and that not even Google gets out clean: "You see real demand, and we are constrained in our ability to serve that demand." "When we go through these investment cycles, there are moments we overshoot -- collectively, as an industry." "We can look back at the internet: there was clearly a lot of excess investment." "It's both rational, and there are elements of irrationality through a moment like this." "No company is going to be immune, including us. If you overinvest, we'll have to work through that phase." The tell isn't that he admitted the irrationality. It's that the man sitting on the most compute on earth is already pricing in the overshoot -- while telling you the demand underneath it is real. A bubble and a revolution were never mutually exclusive. The internet was both.

Sundar Pichai, the CEO of Alphabet, was asked point blank if AI is a bubble. He didn't dodge. He said yes -- partly -- and that not even Google gets out clean: "You see real demand, and we are constrained in our ability to serve that demand." "When we go through these investment cycles, there are moments we overshoot -- collectively, as an industry." "We can look back at the internet: there was clearly a lot of excess investment." "It's both rational, and there are elements of irrationality through a moment like this." "No company is going to be immune, including us. If you overinvest, we'll have to work through that phase." The tell isn't that he admitted the irrationality. It's that the man sitting on the most compute on earth is already pricing in the overshoot -- while telling you the demand underneath it is real. A bubble and a revolution were never mutually exclusive. The internet was both.

Karl Mehta

408,954 views • 1 month ago

$GLW Corning: The AI Optical Infrastructure Springboard Strategy. Investment Thesis. New: 6/29/26. Corning has repositioned a traditional materials manufacturer into a critical supplier at the physical layer of AI infrastructure — a transition that leverages decades of fiber optics and photonics expertise into a market where that expertise is now mission-critical rather than commoditized. High-density fiber and advanced photonics are the connectivity backbone that AI data centers require as networking bottlenecks become increasingly binding constraints on cluster performance, and Corning's manufacturing scale and materials science depth position it as one of the few suppliers capable of meeting that demand at the volumes hyperscalers require. The NVIDIA and Meta partnerships are significant validation points. Both represent demanding customers whose technical requirements and qualification standards are rigorous, and their direct engagement with Corning signals that the company's optical infrastructure capabilities are viewed as strategically necessary rather than substitutable. That kind of direct hyperscaler relationship is difficult for competitors to displace once established, given the integration depth required in data center network architecture planning. The Springboard strategy is the financial framework converting the AI infrastructure opportunity into demonstrated margin improvement. Operating margin expansion that has already materialized provides credibility to the more aggressive 2030 revenue growth targets — this is not a purely forward-looking narrative but one with a track record of execution behind it. That said, the distance between current results and the 2030 targets is substantial, and the growth trajectory assumes continued AI infrastructure capital spending at a pace that has historically been difficult to sustain without periodic digestion phases. Capital intensity is the structural constraint on returns during the buildout phase. Scaling fiber and photonics manufacturing capacity to meet AI-driven demand requires sustained capital deployment, and the return on that investment depends on demand durability matching the capacity being built. Customer concentration compounds that risk — significant revenue exposure to a small number of hyperscaler relationships means that any shift in AI infrastructure capital spending plans at a major customer would disproportionately affect Corning's growth trajectory relative to a more diversified customer base. The Solar business is a complicating factor that sits somewhat apart from the core AI optical infrastructure narrative. Scaling that segment successfully requires different operational capabilities and serves a different demand driver, and management attention split across a capital-intensive solar scale-up alongside the AI infrastructure buildout introduces execution complexity that pure-play AI infrastructure companies don't carry. The valuation reflects multiple years of anticipated growth, which means the premium is justified only if execution stays on pace with the Springboard targets and AI infrastructure capital spending remains robust through the multi-year buildout period the thesis depends on. Corning's market position is genuinely dominant in its core optical infrastructure niche — the question is whether that dominance, expressed through a still-developing financial trajectory, supports a price that has already captured much of the anticipated upside.

$GLW Corning: The AI Optical Infrastructure Springboard Strategy. Investment Thesis. New: 6/29/26. Corning has repositioned a traditional materials manufacturer into a critical supplier at the physical layer of AI infrastructure — a transition that leverages decades of fiber optics and photonics expertise into a market where that expertise is now mission-critical rather than commoditized. High-density fiber and advanced photonics are the connectivity backbone that AI data centers require as networking bottlenecks become increasingly binding constraints on cluster performance, and Corning's manufacturing scale and materials science depth position it as one of the few suppliers capable of meeting that demand at the volumes hyperscalers require. The NVIDIA and Meta partnerships are significant validation points. Both represent demanding customers whose technical requirements and qualification standards are rigorous, and their direct engagement with Corning signals that the company's optical infrastructure capabilities are viewed as strategically necessary rather than substitutable. That kind of direct hyperscaler relationship is difficult for competitors to displace once established, given the integration depth required in data center network architecture planning. The Springboard strategy is the financial framework converting the AI infrastructure opportunity into demonstrated margin improvement. Operating margin expansion that has already materialized provides credibility to the more aggressive 2030 revenue growth targets — this is not a purely forward-looking narrative but one with a track record of execution behind it. That said, the distance between current results and the 2030 targets is substantial, and the growth trajectory assumes continued AI infrastructure capital spending at a pace that has historically been difficult to sustain without periodic digestion phases. Capital intensity is the structural constraint on returns during the buildout phase. Scaling fiber and photonics manufacturing capacity to meet AI-driven demand requires sustained capital deployment, and the return on that investment depends on demand durability matching the capacity being built. Customer concentration compounds that risk — significant revenue exposure to a small number of hyperscaler relationships means that any shift in AI infrastructure capital spending plans at a major customer would disproportionately affect Corning's growth trajectory relative to a more diversified customer base. The Solar business is a complicating factor that sits somewhat apart from the core AI optical infrastructure narrative. Scaling that segment successfully requires different operational capabilities and serves a different demand driver, and management attention split across a capital-intensive solar scale-up alongside the AI infrastructure buildout introduces execution complexity that pure-play AI infrastructure companies don't carry. The valuation reflects multiple years of anticipated growth, which means the premium is justified only if execution stays on pace with the Springboard targets and AI infrastructure capital spending remains robust through the multi-year buildout period the thesis depends on. Corning's market position is genuinely dominant in its core optical infrastructure niche — the question is whether that dominance, expressed through a still-developing financial trajectory, supports a price that has already captured much of the anticipated upside.

TheValueist

16,739 views • 1 month ago

Jeff Bezos was asked about the AI bubble and answered with a story about people who lost everything. Bezos: “This is real. The benefits to society from AI are going to be gigantic.” Society, rather than shareholders. The example he reaches for makes the difference exact. Elkann: “If we go back 25 years ago, when the internet was in that bubblish moment, no one would have predicted a lot of the industrial benefits. And the fiber, huge investments were put in infrastructure.” Bezos agrees, then adds the part nobody remembers. Bezos: “Yes, it’s a perfect example. That all of that fiber optic cable that got laid, and by the way, the companies who laid all that cable went out of business.” Two things happened at the same time. The cable went into the ground and stayed there. The equity went to zero and stayed there. No rational investor would have funded that much fiber. The demand did not exist yet, and the returns came nowhere close to justifying the spend. Mania funded it instead. Overbuilding is the one thing a bubble is reliably good at. Every mile of it is still in the ground. Netflix streams on it. The entire cloud economy runs on capacity bought at a loss by people who were wiped out before the payoff arrived. That is what a bubble actually does. It moves money from investors to the public and calls it a market correction. The datacenters going up now are the same shape. Concrete, power, and silicon that will outlast whoever is currently paying for them. The companies that won the internet built on top of what everyone else paid for. Amazon was one of them.

Jeff Bezos was asked about the AI bubble and answered with a story about people who lost everything. Bezos: “This is real. The benefits to society from AI are going to be gigantic.” Society, rather than shareholders. The example he reaches for makes the difference exact. Elkann: “If we go back 25 years ago, when the internet was in that bubblish moment, no one would have predicted a lot of the industrial benefits. And the fiber, huge investments were put in infrastructure.” Bezos agrees, then adds the part nobody remembers. Bezos: “Yes, it’s a perfect example. That all of that fiber optic cable that got laid, and by the way, the companies who laid all that cable went out of business.” Two things happened at the same time. The cable went into the ground and stayed there. The equity went to zero and stayed there. No rational investor would have funded that much fiber. The demand did not exist yet, and the returns came nowhere close to justifying the spend. Mania funded it instead. Overbuilding is the one thing a bubble is reliably good at. Every mile of it is still in the ground. Netflix streams on it. The entire cloud economy runs on capacity bought at a loss by people who were wiped out before the payoff arrived. That is what a bubble actually does. It moves money from investors to the public and calls it a market correction. The datacenters going up now are the same shape. Concrete, power, and silicon that will outlast whoever is currently paying for them. The companies that won the internet built on top of what everyone else paid for. Amazon was one of them.

Dustin

14,227 views • 8 days ago

The Israeli Ministry of Defense is seeking proposals for a system to counter fiber-optic drones, which are immune to traditional jamming and electronic warfare. The IDF has not been well prepared for Hezbollah’s use of fiber-optic FPVs in Lebanon. 1/

The Israeli Ministry of Defense is seeking proposals for a system to counter fiber-optic drones, which are immune to traditional jamming and electronic warfare. The IDF has not been well prepared for Hezbollah’s use of fiber-optic FPVs in Lebanon. 1/

Roy🇨🇦

83,255 views • 4 months ago

$Jeff Bezos just said an AI crash would be the best thing that ever happened to AI. Sounds insane until you look at what happened last time. Telecom companies invested $500 billion into fiber optic cable between 1996 and 2001. Most of it financed with debt. The growth in capacity outstripped demand by orders of magnitude. By 2005, 85% of the fiber they laid was still dark. Unused. The companies that built it went bankrupt. WorldCom. Global Crossing. 360networks. Exposed as fraud, overleveraged, or both. $2 trillion in telecom stock value evaporated. The cable stayed in the ground. 80.2 million miles of fiber, representing 76% of all digital wiring in the United States at that point, survived every bankruptcy filing and liquidation sale. The glass in the ground didn't care who owned the company above it. Amazon launched Prime Video on that fiber. Netflix streamed its first original on it. Google built YouTube on bandwidth that cost 90% less than it would have before the crash because dead companies had already paid for the capacity. Every cloud platform, every streaming service, every video call you've taken in the last decade runs on infrastructure that bankrupt telecom companies subsidized with investor money they never returned. Now look at 2026. Big Tech is spending $400 billion on data centers this year alone. Amazon's capex plan is $200 billion, the largest in corporate history. The AI bubble, by one estimate, is 17 times the size of the dot-com bubble. Bezos himself called it an "industrial bubble" at Italian Tech Week. Here's why he's calm about it. He watched the exact same movie in 2001. Amazon nearly died in that crash. Stock dropped 90%. But when the dust settled, Bezos had cheap bandwidth, empty data centers looking for tenants, and a trained workforce of engineers that the dead dot-coms had paid to train. AWS was built on the rubble. The pattern is specific. Investors fund infrastructure. Companies die. Infrastructure survives. The next generation builds on it at a fraction of the original cost. Data centers don't disappear when an AI startup folds. Chips depreciate, but the buildings, the power connections, the cooling systems, the land with energy contracts attached to it: all of that persists. When AI companies fail, they'll leave behind computing capacity that someone else will rent for pennies on the dollar. Bezos is making a $200 billion bet with full awareness that much of the industry around him will collapse. He's seen this before. The last time the bubble popped, it handed him the raw materials to build a $2 trillion company. The investors who funded the fiber never saw a return. The company that inherited it became the most valuable on earth.$

Jeff Bezos just said an AI crash would be the best thing that ever happened to AI. Sounds insane until you look at what happened last time. Telecom companies invested $500 billion into fiber optic cable between 1996 and 2001. Most of it financed with debt. The growth in capacity outstripped demand by orders of magnitude. By 2005, 85% of the fiber they laid was still dark. Unused. The companies that built it went bankrupt. WorldCom. Global Crossing. 360networks. Exposed as fraud, overleveraged, or both. $2 trillion in telecom stock value evaporated. The cable stayed in the ground. 80.2 million miles of fiber, representing 76% of all digital wiring in the United States at that point, survived every bankruptcy filing and liquidation sale. The glass in the ground didn't care who owned the company above it. Amazon launched Prime Video on that fiber. Netflix streamed its first original on it. Google built YouTube on bandwidth that cost 90% less than it would have before the crash because dead companies had already paid for the capacity. Every cloud platform, every streaming service, every video call you've taken in the last decade runs on infrastructure that bankrupt telecom companies subsidized with investor money they never returned. Now look at 2026. Big Tech is spending $400 billion on data centers this year alone. Amazon's capex plan is $200 billion, the largest in corporate history. The AI bubble, by one estimate, is 17 times the size of the dot-com bubble. Bezos himself called it an "industrial bubble" at Italian Tech Week. Here's why he's calm about it. He watched the exact same movie in 2001. Amazon nearly died in that crash. Stock dropped 90%. But when the dust settled, Bezos had cheap bandwidth, empty data centers looking for tenants, and a trained workforce of engineers that the dead dot-coms had paid to train. AWS was built on the rubble. The pattern is specific. Investors fund infrastructure. Companies die. Infrastructure survives. The next generation builds on it at a fraction of the original cost. Data centers don't disappear when an AI startup folds. Chips depreciate, but the buildings, the power connections, the cooling systems, the land with energy contracts attached to it: all of that persists. When AI companies fail, they'll leave behind computing capacity that someone else will rent for pennies on the dollar. Bezos is making a $200 billion bet with full awareness that much of the industry around him will collapse. He's seen this before. The last time the bubble popped, it handed him the raw materials to build a $2 trillion company. The investors who funded the fiber never saw a return. The company that inherited it became the most valuable on earth.

Aakash Gupta

104,079 views • 4 months ago

🦔 Bezos says AI is an "industrial bubble" not a "financial bubble," and that's supposed to make it okay. His argument: investors may lose money, but society still gets the inventions. Like the dot-com bust left us with the internet infrastructure. He's not wrong about the pattern. The dot-com crash left behind dark fiber that companies still use today. But notice the framing shift. We went from "it's not a bubble" to "it's a bubble, but it's the good kind." That's progress, I guess. My Take The guy with billions riding on AI telling you the AI bubble is actually fine is worth taking with some skepticism. Yes, some useful technology will survive whatever correction comes. But trillions in investment requires more than "useful." It requires world-changing returns to justify the spending. The real question is who pays for the difference between what was invested and what actually pays off. In the dot-com crash, retail investors and pension funds took massive losses. The infrastructure survived, but people's retirement accounts didn't. Bezos emerged from that wreckage as one of the richest people alive. When billionaires tell you bubbles can benefit society, ask who specifically benefits and who holds the bag. Hedgie🤗

🦔 Bezos says AI is an "industrial bubble" not a "financial bubble," and that's supposed to make it okay. His argument: investors may lose money, but society still gets the inventions. Like the dot-com bust left us with the internet infrastructure. He's not wrong about the pattern. The dot-com crash left behind dark fiber that companies still use today. But notice the framing shift. We went from "it's not a bubble" to "it's a bubble, but it's the good kind." That's progress, I guess. My Take The guy with billions riding on AI telling you the AI bubble is actually fine is worth taking with some skepticism. Yes, some useful technology will survive whatever correction comes. But trillions in investment requires more than "useful." It requires world-changing returns to justify the spending. The real question is who pays for the difference between what was invested and what actually pays off. In the dot-com crash, retail investors and pension funds took massive losses. The infrastructure survived, but people's retirement accounts didn't. Bezos emerged from that wreckage as one of the richest people alive. When billionaires tell you bubbles can benefit society, ask who specifically benefits and who holds the bag. Hedgie🤗

Hedgie

18,818 views • 7 months ago

Jeff Bezos built a trillion dollar company on infrastructure paid for by people who went bankrupt. And he just explained why it’s about to happen again with AI. In the 1990s, companies poured billions into laying fiber optic cable across the planet. Every one of them went broke. Bezos: “All of that fiber optic cable that got laid, and by the way, the companies who laid all that cable went out of business.” They died. The cable stayed. And it became the foundation of everything that followed. Now it’s happening again. Six people. No product. They started yesterday. $20 billion valuation. Bezos: “Investors don’t usually give a team of 6 people a couple of billion dollars with no product. It’s rare. And that’s happening today.” Everyone sees this and screams bubble. Bezos sees the same thing and recognizes something older. There are two kinds of bubbles. A financial bubble, like 2008, is pure destruction. Capital vanishes. Society inherits the crater. An industrial bubble runs on entirely different logic. Bezos: “This is a kind of industrial bubble, as opposed to financial bubbles.” The 90s biotech boom burned through billions. Stocks collapsed. Companies disappeared. The drugs they created are still saving lives today. Bezos: “As a group, they all lost money. But we did get a couple of life saving drugs.” The investors lose. The inventions remain. Every single time. Most AI companies being funded right now won’t exist in five years. The people writing the checks already know it. They are not confused. They are not reckless. They are financing infrastructure the entire world will inherit. Bezos: “The benefits to society from AI are going to be gigantic.” 25 years ago, no one could have predicted what the internet would become. Least of all the people who laid the cable. They went broke before the world they built ever switched on. The AI bubble is not a sign that something has gone wrong. It is how civilization has always financed things too large for any rational person to justify. Every piece of technology you use without thinking was once someone’s life savings disappearing in real time. That is what a successful bubble looks like from the other side. You stop seeing it. It just becomes the world. The system only works because the builders believe they will be the exception. If they ever saw the pattern clearly, they would stop. And the future would stop arriving. Progress does not run on genius. It runs on brilliant people being catastrophically, necessarily wrong. The people who build the future have never been the ones who get to live in it. That is not a flaw in the system. That is the system.

Jeff Bezos built a trillion dollar company on infrastructure paid for by people who went bankrupt. And he just explained why it’s about to happen again with AI. In the 1990s, companies poured billions into laying fiber optic cable across the planet. Every one of them went broke. Bezos: “All of that fiber optic cable that got laid, and by the way, the companies who laid all that cable went out of business.” They died. The cable stayed. And it became the foundation of everything that followed. Now it’s happening again. Six people. No product. They started yesterday. $20 billion valuation. Bezos: “Investors don’t usually give a team of 6 people a couple of billion dollars with no product. It’s rare. And that’s happening today.” Everyone sees this and screams bubble. Bezos sees the same thing and recognizes something older. There are two kinds of bubbles. A financial bubble, like 2008, is pure destruction. Capital vanishes. Society inherits the crater. An industrial bubble runs on entirely different logic. Bezos: “This is a kind of industrial bubble, as opposed to financial bubbles.” The 90s biotech boom burned through billions. Stocks collapsed. Companies disappeared. The drugs they created are still saving lives today. Bezos: “As a group, they all lost money. But we did get a couple of life saving drugs.” The investors lose. The inventions remain. Every single time. Most AI companies being funded right now won’t exist in five years. The people writing the checks already know it. They are not confused. They are not reckless. They are financing infrastructure the entire world will inherit. Bezos: “The benefits to society from AI are going to be gigantic.” 25 years ago, no one could have predicted what the internet would become. Least of all the people who laid the cable. They went broke before the world they built ever switched on. The AI bubble is not a sign that something has gone wrong. It is how civilization has always financed things too large for any rational person to justify. Every piece of technology you use without thinking was once someone’s life savings disappearing in real time. That is what a successful bubble looks like from the other side. You stop seeing it. It just becomes the world. The system only works because the builders believe they will be the exception. If they ever saw the pattern clearly, they would stop. And the future would stop arriving. Progress does not run on genius. It runs on brilliant people being catastrophically, necessarily wrong. The people who build the future have never been the ones who get to live in it. That is not a flaw in the system. That is the system.

Dustin

84,842 views • 1 month ago

.Ben Shapiro: Are you worried about an AI bubble similar to the dot-com bubble? Anson: "At some point, there might be some sort of a bubble, but every single bubble that we've had in the United States has led to, again in the longer term, more prosperity."

.Ben Shapiro: Are you worried about an AI bubble similar to the dot-com bubble? Anson: "At some point, there might be some sort of a bubble, but every single bubble that we've had in the United States has led to, again in the longer term, more prosperity."

Daily Wire

77,969 views • 20 days ago