Who is on this subjectordered by the date of their first quote here
2 of 8 lanes rest on fewer than 5 quotes and are marked thin. Offsets are days from the middle first-quote date, 11 Jun 2024 — a date, and nothing else. It is not a claim about who reached a view first.
-
David Friedberg
6
2 Oct 2023 – 18 Jul 2026
-253 d
- Dylan Patel 32 5 Dec 2023 – 25 Aug 2026 -189 d
-
Bill Gurley
11
4 Apr 2024 – 11 May 2026
-68 d
-
Brad Gerstner
6
11 Jun 2024 – 15 May 2026
+0 d
-
Gavin Baker
1
thin
27 Aug 2024 – 27 Aug 2024
+77 d
- Jensen Huang 8 13 Oct 2024 – 16 Jun 2026 +124 d
- Sir Richard Dearlove 6 10 Jun 2025 – 16 Oct 2025 +364 d
-
Marc Rowan
2
thin
3 Mar 2026 – 3 Mar 2026
+630 d
The chronologysourced and dated, oldest first
-
-
“In fact, we did federal Freedom of Information Act requests for the federal government in The U. S. To get hard drives and hard drives of data.”
-
Dylan Patel
Patel argues AI software will have lower R&D costs but much higher cost of goods sold from operating services.
“Like the R and D cost is much lower in terms of people, but the cost of goods sold in terms of actually operating the service, I think will be much higher.”
-
-
-
Bill Gurley
Gurley cites Microsoft opening a data center every three days, with data centers consuming nearly 10% of US electricity.
“Microsoft alone's opening opening a new data center globally every three days. These power hungry operations will together consume more than 480 terawatts hours of electricity or almost a tenth of US demand.”
-
Bill Gurley
Gurley says China has massive advantage clearing regulatory decks permanently while US only grants exceptions per project.
“This is where I think China has a massive advantage. They could just clear the decks for good for every project going forward. Whereas we're now making exceptions maybe with the TSMC plant in Arizona.”
-
Brad Gerstner
Gerstner says AI customers must place orders five years in advance and secure power and data centers.
“they have to place orders five years in advance for these products. They have to go find power. They have to find data centers”
-
Gavin Baker
Baker says GPUs got 50x faster while data centers only 4-5x faster, creating investment opportunity in infrastructure.
“GPUs have gotten 50 times faster, and the rest of the data center has only gotten four to five times faster.”
-
Dylan Patel
Patel claims China could build a gigawatt data center in six months near the Three Gorges Dam but hasn't.
“China could just build it in six months, I think, around the 3 Gorges Dam or many other places. Right?”
-
Dylan Patel
Patel says Microsoft and OpenAI have 500,000 GB200 GPUs totaling one gigawatt coming online next year.
“Microsoft slash OpenAI slash their partners for them. And then and then potentially even more, 500 k GB two hundreds, right, is a gigawatt. Right? And that's, like, online next year.”
-
Dylan Patel
Patel predicts OpenAI will raise 50 to 100 billion dollars by early next year to fund planned cluster builds.
“there's no fucking way you can pay for the scale of clusters that are being planned to be built next year for OpenAI until unless they raise, like, 50 to $100,000,000,000, which I think they will raise that, like, end of this year, early next year.”
-
Jensen Huang
Huang says traditional supercomputers take three years to plan plus one year to get working.
“A supercomputer that you would build would take normally three years to plan. Right. And then they deliver the equipment and it takes one year to get it all working.”
-
Jensen Huang
Huang predicts a new layer of AI factories running digital workers continuously.
“there's gonna be a new infrastructure. This new infrastructure are going to be AI factories that operate these digital humans. And they're gonna be running all the time, 20 fourseven.”
-
Dylan Patel
Patel reports Microsoft plans 1.5 gigawatts in one site within two and a half years across four 300 megawatt buildings.
“And so the scale that they wanna reach to, 1.5 gigawatts in two and a half years in one site.”
-
Bill Gurley
Gurley draws parallel to internet era when companies migrated away from Oracle and Sun as optimization replaced development.
“And five years later, they weren't on Oracle or Sun. And some have argued it went from a development sandbox world to a optimization world.”
-
Bill Gurley
Patel argues scaling continues because hyperscalers are building multi-gigawatt data centers and connecting them with billions in fiber purchases.
“Why is why is Amazon building these multi gigawatt data centers? Why is Google? Why is Microsoft building multiple gigawatt data centers plus buying billions and billions of dollars of fiber to connect them together”
-
Bill Gurley
Patel argues hyperscalers are building multi-gigawatt data centers and buying billions in fiber to win on scale
“Why is Microsoft building multiple gigawatt data centers plus buying billions and billions of dollars of fiber to connect them together because they think, hey.”
-
-
-
Bill Gurley
Gurley calls anti-Chinese sentiment irrational given China's lead in EVs and infrastructure while America maintains false superiority.
“I think this anti Chinese mentality is a little nuts because if you look at low cost EVs, if you look at their subway stations, if you look at, they're leading us in many areas.”
-
Dylan Patel
Patel breaks down full H100 GPU cost at $40,000 to $45,000 including networking and infrastructure versus $24,000 for chip alone.
“each GPU, right, once you include networking, building, all this sort of stuff maybe is you know, the GPU itself of h 100 is, like, 24,000, but once you add everything else up, it's, like, forty, forty five thousand dollars per GPU all in of everything.”
-
Dylan Patel
Patel says in some Midwest regions, grid transmission costs exceed power generation costs, constraining data center buildouts.
“It's like, what the flip? Right? Like, it's like so, like, the grid needs huge investments.”
-
Dylan Patel
Patel forecasts data center power will grow from 150 megawatts today to gigawatt-scale clusters by 2026.
“We're seeing, like, you know, 400, 500 megawatts cluster sort of being built out now. Right? Like you know? And and then, you know, in 2026, I'd imagine it'll be, like, gigawatt scale clusters.”
-
Dylan Patel
Patel identifies a 40 gigawatt power shortfall for US data centers even with all planned capacity additions.
“There is an over 40 gigawatt hole in power for The US. Even if we don't turn off any more coal, in fact we restart some, we take all the natural gas dual combined cycle reactor deliveries that we're going to have, and you add in renewables, you have a 40 gigawatt hole of power production in The US.”
-
Dylan Patel
Patel identifies a 40 gigawatt power deficit in US even with all planned gas and restarted coal.
“Even if we don't turn off any more coal, in fact we restart some, we take all the natural gas dual combined cycle reactor deliveries that we're going to have,”
-
Dylan Patel
Patel says certain states will see over half their power consumed by AI within a few years.
“When we look at power consumption by state, something pretty simple, there are certain states where more than half their power is going be consumed by AI just in the next few years.”
-
Dylan Patel
Patel argues that efficiency gains alone without capability increases cannot justify massive AI infrastructure investments.
“That would not pay for all of these build outs. Right? AI is useful today, but it's not capable of doing a lot of things.”
-
David Friedberg
Friedberg recalls Google building shipping container data centers twenty years ago requiring only water and power.
“Urs and his team were building these super secret shipping container data centers. They had these like data center in a box that you could ship anywhere.”
-
David Friedberg
Friedberg recalls Google built shipping container data centers twenty years ago that only needed water and power.
“As long as you had access to water and power, it could connect to the internet and you could scale data center capacity all over the world. That was twenty years ago.”
-
Dylan Patel
Patel notes CoreWeave has fully liquid cooled data centers while hyperscalers are still working on designs.
“They're doing full liquid cooling as well and they're working on those designs, but, you guys have fully liquid cooled data centers.”
-
Dylan Patel
Patel says CoreWeave can build $10 billion clusters from start to completion much faster than competitors.
“it lets you get to these, you know, call it $10,000,000,000 clusters, right, in in time scales, from project start to completion that are much much shorter than anyone else.”
-
Dylan Patel
Patel describes customer with over 10,000 GPUs whose engineers demanded CoreWeave infrastructure after acquisition.
“What's what's funny is there was a customer that had more than 10,000 GPUs with you. They got acquired.”
-
Sir Richard Dearlove
Dearlove warns Chinese cars have cellular modules that could be reprogrammed remotely to immobilize vehicles in a crisis.
“Because in every Chinese car there is a cellular module which is connected to the internet of things and that car in a crisis can be reprogrammed by the manufacturer in China.”
-
Sir Richard Dearlove
Dearlove warns Chinese cars can be remotely immobilized from Beijing, potentially paralyzing UK cities in crisis.
“that car in a crisis can be reprogrammed by the manufacturer in China. And you can find your Chinese car, 40 of them, let's say around London immobilized.”
-
Sir Richard Dearlove
Dearlove warns China could immobilize Chinese cars remotely and bring UK cities to a halt in a crisis.
“Chinese could bring in a crisis UK city to a grinding halt, bang, with a flick of a switch in Beijing.”
-
Sir Richard Dearlove
Dearlove says Chinese-made renewable energy equipment with cellular modules could be remotely disabled from China.
“A lot of our renewable sources of energy are Chinese manufactured with cellular modules in them that can be reprogrammed from China. You could knock out the whole of a North Sea wind farm, bang.”
-
Dylan Patel
Patel reveals Meta is building temporary tent structures to house GPUs because permanent buildings take too long to construct.
“Meta is literally building these, like, temporary, like, tent structures to put GPUs in because building the building takes too long, and it takes too much labor, as you mentioned labor.”
-
Dylan Patel
Patel describes former Yandex workers paid bonuses for speed who use drugs to complete data center buildouts faster.
“Like and they get paid bonuses for being faster, and therefore, do, like, certain drugs to be able to finish the build outs faster.”
-
Dylan Patel
Patel states 80% of GPU data center cost is capital equipment, only 20% is land, power, and cooling.
“It's the it's the physical data center conversion power conversion equipment. All of this stuff is, like, 80% of the cost. And then 20% is gonna be your land and your power and your cooling”
-
Dylan Patel
Patel reports new Trump tax bill allows year-one GPU depreciation, worth $10 billion annually to Meta alone.
“the new the new Trump, you know, tax bill institutes something really incredible, which is that you can depreciate all of the GPU cluster cost in year one, which we put out, like, a note about how, like, the tax implications to, like, meta are, like, $10,000,000,000 a year. And across each of the major hyperscalers, it's, like, massive.”
-
Brad Gerstner
Gerstner highlights US fission reactors cost four to five times more and none are being built.
“Why does it cost us four to five times to build a fission reactor in this country? Why are we building no nuclear reactors in this country?”
-
Dylan Patel
Patel contrasts traditional data centers with minimal retrofit costs against modern GPU deployments requiring major upgrades each generation.
“it used to be you'd build a data center and there wouldn't really be much retrofits cost, right? You would slide servers in and out, you would rent racks to people.”
-
Dylan Patel
Patel describes Meta's temporary tent-like data centers designed for single GPU generation lifecycles instead of 30-year buildings.
“Meta has got a data center design that's only built to last for a single generation. They put them in these tent buildings.”
-
Sir Richard Dearlove
Dearlove says the proposed Chinese embassy site is sensitive due to communications cables running underneath or near it.
“And the reason it's sensitive in my understanding is because of the communications cables that run underneath it or very close to it.”
-
Sir Richard Dearlove
Dearlove states UK is in state of grey warfare with Russia using sub-conflict means against infrastructure.
“You know as far as the Russians are concerned because of Ukraine we're in a state of grey warfare with them and they are using all sorts of means which are beneath the level of open conflict to cause, well, mayhem in aspects of the West's daily critical infrastructure, daily life,”
-
Dylan Patel
Patel quantifies InferenceMax uses tens of millions of dollars in GPU hardware with daily software updates.
“we have tens of millions of dollars of hardware of the current and last generation GPUs. As I mentioned before, the software updates every single day.”
-
Bill Gurley
Gurley states China builds nuclear fission plants at one fourth the price of US construction costs.
“So they are building new nuclear fission plants. So fission being old school, not new school, at one fourth the price that we do it here in The US.”
-
Bill Gurley
Gurley reports China builds nuclear fission plants at one-fourth the US price; South Korea does too.
“So fission being old school, not new school, at one fourth the price that we do it here in The US. So is South Korea, by the way.”
-
David Friedberg
Friedberg states China is constructing 400 nuclear reactors representing 60% of total US power production capacity.
“Just try to think about each one of those being one, maybe one and a half gigawatts. Suddenly, that's 60% of The US power production capacity being built right now”
-
David Friedberg
Friedberg notes China is building 400 nuclear reactors, equivalent to 60% of total US power production capacity.
“Just try to think about each one of those being one, maybe one and a half gigawatts. Suddenly, that's 60% of The US power production capacity being built right now in just nuclear by China.”
-
Brad Gerstner
Gerstner says $250 billion on NVIDIA implies $500 billion total AI CapEx spending.
“one of the things that keeps investors like me up in the nights is a 250,000,000,000 being spent on NVIDIA on on the semi side of things.”
-
-
-
“In 2023, these companies spent $150,000,000,000 on capex building out data centers. This year, over 500,000,000,000. Though, that's not speculative. That is purchase orders. Those are buildings. That's power.”
-
Dylan Patel
Patel explains agentic training only needs to sync tokens every few minutes versus weights every seconds in pre-training.
“When you're doing these rollouts and especially as things get more and more agentic and training, you might not only need to send not the entire weights but just the tokens that are relevant.”
-
Dylan Patel
Patel notes semiconductor industry doubles transistors yearly but energy industry wasn't prepared for similar growth.
“Whereas the energy industry in America wasn't. And and so, like, initially, people were, like, not creative. They're like, let's do let's do these kinds of gas plants.”
-
Jensen Huang
Huang says the entire computing stack—processing, storage, networking, security—is being reinvented for AI.
“But remember what computing is. There's computing, there's the processing, but there's storage, networking, and security. All that is being reinvented as we speak.”
-
Jensen Huang
Huang says three formerly separate industries are converging rapidly to build AI infrastructure.
“Right now, as we're speaking, these three entirely what used to be three different industries are all growing incredibly fast so that we could create the infrastructure for intelligence and manufacturing the AIs.”
-
Jensen Huang
Huang states a gigawatt AI factory costs approximately fifty billion dollars and tens are being built worldwide.
“While these factories are incredibly complex, they're you know, a gigawatt AI factory is about $50,000,000,000 And now we're building tens of gigawatts around the world.”
-
Jensen Huang
Huang characterizes current AI infrastructure as the largest industrial buildout in human history.
“It's an enormous infrastructure build out, the largest industrial infrastructure build out in human history.”
-
Jensen Huang
Huang cites estimates of eighty-five trillion dollars in infrastructure investment over the next ten years.
“Trillions of dollars, tens of trillions of dollars. Estimates have it close to $100,000,000,000,000 $85,000,000,000,000 in the next ten years.”
-
Dylan Patel
Patel calculates $100B AI revenue requires $250B infrastructure with five-year depreciation at 50% margins.
“That $50,000,000,000 of COGS needs to burn on infra, which cost roughly with if a five if you're talking about five year depreciation, call it $250,000,000,000,”
-
Dylan Patel
Patel calculates $100B AI revenue requires $250B infrastructure spend at five-year depreciation, double current capex.
“That $50,000,000,000 of COGS needs to burn on infra, which cost roughly with if a five if you're talking about five year depreciation, call it $250,000,000,000, right, of infra Yeah. For a $100,000,000,000 of revenue.”
-
Brad Gerstner
Gerstner compares the next three years of AI investment to building the interstate highway system, all privately funded.
“I think for the next three years, this is like building the interstate highway system. Right? This is significant. It's important to The United States Of America, by the way.”
-
Bill Gurley
Gurley cites UK, China, India, and Brazil all created instant digital payment systems years ago.
“The UK created something called Faster Payment seventeen years ago. And China has one and India has one. Brazil created one seven years ago, six years ago called PIX that's wildly successful.”
-
Dylan Patel
Patel says Google spending $180B and Amazon $200B on AI infrastructure represents 4x increase from recent years.
“If we start looking at like, hey, this year, Google spending 2 Amazon spending $200,000,000,000, Google spending a $180,000,000,000 on on AI infrastructure primarily. Right?”
-
Marc Rowan
Rowan asserts the world is simultaneously building infrastructure, energy, manufacturing, defense, and AI, creating the largest capital need ever.
“S, they want to do in Europe, and they're not as well prepared to it, and they're doing it in The Middle East, and they're doing it elsewhere around the world.”
-
Marc Rowan
Rowan describes unprecedented simultaneous global infrastructure spending across US, Europe, and Middle East for AI, energy, and manufacturing.
“S, they want to do in Europe and they're not as well prepared to it and they're doing it in The Middle East and they're doing it elsewhere around the world.”
-
Bill Gurley
Gurley says it costs 400% more to build nuclear plants in US than in South Korea or China.
“it costs four x, 400% more to build a a a fission nuclear plant, 70 year old technology in The US than it does in South Korea or China.”
-
Bill Gurley
Gurley states South Korea builds nuclear plants at one fourth the US cost, calling it alarming.
“South Korea, not China, South Korea can build a nuclear fission plant for one fourth the price that we can in The US. And someone needs to be massively alarmed by that.”
-
Brad Gerstner
Gerstner warns that 30-40% of compute is delayed this year, which could slow the entire AI trade if delays persist.
“And you heard the rumors and seen the headlines that 30 or 40% of compute now is delayed this year. So keep your eye on power and compute.”
-
Jensen Huang
Huang forecasts tens of trillions of dollars in new computer infrastructure over the next decade.
“We're now in the process of reinventing all of that, which is the reason why over the course of the next ten years, we're going to be building tens of trillions of dollars of new computers to replace the old computers that we built over the last sixty years.”
-
Dylan Patel
Patel forecasts data center capacity will reach 30 gigawatts in the next year.
“This year, we're deploying 20 gigawatts of data centers. Next year, that number goes up 50 gigawatts or 50%, sorry. 50%, sorry.”
-
Dylan Patel
Patel says a data center is physically redoing doors to accommodate the double-width Helios racks.
“There's a there's a data center that I know of specifically that is going to be deploying Helios, and they are redoing the doors so that they can reel the Helios in.”
-
“there was a poll that came out today, something north of 50% of Americans believe that data centers increase the cost of water and electricity.”
-
Dylan Patel
Patel forecasts AI CapEx will exceed $2 trillion by 2028, up from just over $1 trillion in 2025.
“As we go forward into the future, the the numbers for computer ballooning, right, we're at, you know, you know, a little bit over a trillion dollars of CapEx this year.”
-
Dylan Patel
Patel estimates $3-4 trillion total CapEx needed by 2028 across compute, data centers, and energy infrastructure.
“So to enable, let's say, that 100 gigawatts by 2030 or let's even like let's even like pare it down to 2028 where it's like 3 or $4,000,000,000,000 of CapEx across all of these items.”
-
Dylan Patel
Patel questions where $3-4 trillion in CapEx will come from for 2028 AI infrastructure buildout.
“So if you're at 3 or $4,000,000,000,000 of CapEx, where does all this cash come from?”
-
Where they overlapsame programme, same subject
- BG2 Pod also carries Bill Gurley, Brad Gerstner and Jensen Huang on this subject, at different times.
- Invest Like the Best also carries Bill Gurley and Gavin Baker on this subject, at different times.