Try Pro free
Full episode 40:16

The best 16 min of 40

My Brutally Honest Thoughts About the AI Bubble.

Steve Eisman says AI can transform the economy, but concentrated spending, weak moats and financing risks could undermine the investment boom.

The Real Eisman Playbook · Steve Eisman

The best minutes

  1. 24:25AI spending depends on a few model companies8 min
  2. 17:01AI infrastructure is turning cash-rich firms into borrowers6 min
  3. 33:58Low switching costs leave AI models exposed to price warsPro pick 3 min

Also worth watching

  1. 5:57Eisman assigns the middle AI scenario the highest probability4 min
  2. 10:22AI exposure reaches far beyond technology stocks3 min
  3. 13:28A 50% software decline can wipe out buyout equity3 min

Skip the opening AI definitions and infrastructure setup, the power-and-permitting discussion, and the closing recap and Substack promotion.

Full analysis · 6 min read

Real Alpha · Nov 9

Steve Eisman says his middle AI scenario has a 40% chance

  • Steve Eisman gives his middle AI scenario a 40% chance, versus 30% each for bull and bear.
  • His concern is not AI's usefulness, but concentrated demand, heavier financing, construction delays, and weak model switching costs.
  • Founders should test which customers and contracts support infrastructure commitments if model demand slows.

Free with an account

Read the rest of the analysis

6 sections, the 5 other voices it tests Eisman against, and what it means for you. Every episode we cut gets one.

Signed up before? The same buttons sign you in. The episode, its moments and the transcript stay free.

8:51

Eisman's 40% middle case leads his scenario tree

Steve Eisman gives the intermediate scenario a 40% chance, versus 30% each for bull and bear. His highest-probability case is not a soft landing.

He describes overbuilding, construction delays, Chinese competition, and a possible price war leaving spending ahead of demand. Brad Saffalo of PAA Research estimates U.S. data-center capacity could reach twice demand by end-2027.

That is Saffalo's estimate, not a measured current surplus. Eisman says that if it proves right, AI capital spending could slow in 2028.

Eisman says the middle case would unwind much of the market's AI gains and bring a bear market for a time. His bear case goes further, with a recession and market crash.

For example, he points to the internet boom. The technology transformed the world, but many first-generation internet companies failed.

“I give this narrative a 40% chance.”
8:51

Our read

The 40% figure captures Eisman's uncertainty, not a timed prediction. His central test is whether spending outruns demand before the technology's benefits spread.

26:18

A few model firms connect AI spending to the economy

Eisman traces risk from Nvidia through hyperscalers to OpenAI and Anthropic. Nvidia's revenue grew more than 100%, but 70% of its accounts receivable came from five customers.

Eisman called receivables a proxy for revenue. He said he could not identify those customers and assumed they were hyperscalers or adjacent companies.

Wall Street estimates he cited put something like 70% of hyperscaler AI revenue with OpenAI and Anthropic. Eisman translated that into 25% to 35% of total cloud revenue, while saying he could not confirm the figures.

Oracle's backlog is around $600 billion, and Eisman said about 50% came from OpenAI. He said that OpenAI share was public information. He also said OpenAI represented over 20% of CoreWeave's contracted order book.

Eisman says hyperscalers are spending about $850 billion in CapEx this year, equal to around half of projected 2% U.S. GDP growth. Winfield Sickles on Inside Blackstone in October 2026 estimated spending above $800 billion this year, probably over $1 trillion next year.

Sickles pointed to cloud revenue growth as evidence that spending is appearing in results. But revenue growth alone does not show whether returns justify the buildout.

Our read

The concentration is a credible transmission risk, but parts of Eisman's chain rely on estimates he cannot verify. Cloud demand supports the bull case without settling profitability.

18:53

AI financing shifts cash-rich firms toward capital markets

Eisman says AI has pushed cash-generating cloud firms from capital-light operations toward outside financing. Oracle's debt rose from $94 billion to $155 billion over two years.

S&P downgraded Oracle to BBB minus, one level above junk. That adds financing risk, though Eisman says leverage alone would not sink the AI story if the investments succeed.

Meta's Louisiana data-center project shows how obligations can sit outside reported debt. The $27 billion project uses an off-balance-sheet vehicle in which Meta owns 20%.

Meta agreed to rent the data center for 20 years and bears project delays and overruns. Ernst & Young flagged the arrangement as a critical audit matter in Meta's 2025 10-K.

Eisman read the auditor's flag as a sign of concern, while acknowledging that his plain-English translation was his own. He says such financing makes the story more dangerous, but not enough by itself to sink it.

Ed Elson on Prof G Markets in November 2026 also focused on hyperscalers' reliance on capital markets. Dylan Patel on Big Technology Podcast in October 2026 highlighted Oracle's fixed-margin OpenAI contracts and GPU price exposure.

Our read

The balance-sheet shift matters even before a company fails. Delays and contract economics can raise funding pressure, so reported debt is only part of the exposure.

29:02

OpenAI's reported numbers make failure a live scenario

OpenAI is the specific weak link Eisman worries about, not a company he says has already failed. Citing Wall Street Journal reporting, he said June-quarter revenue grew 18% from March.

Eisman said revenue rose $1 billion in three months, while costs rose $3 billion. By his account, costs were increasing faster than revenue.

By comparison, he cited Anthropic's June-quarter revenue above $11 billion, up more than 100% from March. These are reported comparisons, not audited accounts presented in the episode.

Eisman says OpenAI first pursued consumer adoption, offering its product free or below cost in hopes of raising prices later. He says the company is now chasing business customers, where Anthropic started earlier.

Eisman relayed Ed Zitron's statement that OpenAI will have to raise $100 billion per year for the next several years to survive. Still, Eisman stopped short of predicting failure and said it was too early.

Timing matters in his scenario. A near-term failure could prompt spending cuts before AI demand becomes more diversified.

Our read

The reported gap makes OpenAI's funding needs a real diligence question, not proof of insolvency. A recession scenario depends on how customers, lenders, and hyperscalers respond.

34:15

Cheap substitutes test whether model pricing can hold

Eisman sees low customer switching barriers as the deepest business-model risk. Useful products do not guarantee durable pricing when customers can move among ChatGPT, Claude, Gemini, and Chinese alternatives.

He put OpenAI and Anthropic flagship prices at $2 to $10 per 1 million tokens. He said open-weight models cost around $0.20 per 1 million tokens.

Our calculation from those figures puts open-weight pricing at roughly 2% to 10% of the frontier range. Eisman says the leading models are superior, but many tasks may not need their highest capability.

Garrison Lovely on Future of Life Institute Podcast in October 2026 also identified open-weight models as a frontier-margin risk. He said they could become good enough for many tasks, not that parity has arrived.

For example, Eisman said Uber used its entire AI budget in a few months after prices rose. He said enterprises then became more cost-conscious and token-maxing ended.

Eisman said reports showed both leading providers cutting prices amid competition from cheaper models. That makes a price war a risk to monitor, not a settled outcome.

“and the switching cost between models seems to be perhaps zero, but certainly very low.”
34:15

Our read

The price gap makes substitution economically meaningful, even if models differ in capability. Usage growth alone cannot show whether providers can cover their costs.

37:30

Infrastructure exposure avoids choosing a model winner

Eisman says he is more interested in infrastructure suppliers because they need not predict which LLM wins. He names Nvidia, Micron, Cisco, GE Vernova, and Eaton.

That framing avoids choosing a model provider, but not exposure to delayed projects, power constraints, or customer cutbacks. Eisman's own scenarios tie infrastructure spending to model-company health.

Andy Jassy said on a16z Podcast in October 2026 that Amazon expected $220 billion in CapEx this year while leaving meaningful demand unmet. That is one company's account of demand, not a guarantee of supplier returns.

A founder can apply the same test internally. For each AI dependency, identify the customer, contract duration, cancellation rights, price path, and funding source.

Then compare incremental revenue or saved labor with serving and infrastructure costs. Usage and contracted demand are inputs, not proof of profitable utilization.

Our read

Infrastructure may reduce dependence on any single model, but it cannot remove exposure to the buildout cycle. The distinction is between demand for capacity and durable returns on that capacity.

What it means for you

  1. Map model exposure. Measure how much revenue, backlog, and capacity commitments depend on one model provider or customer. Check contract duration and cancellation rights alongside headline backlog.
  2. Stress test funding. List debt, equity raises, SPVs, leases, and overrun obligations behind AI infrastructure. Model the cash impact if construction slips or a major customer cuts orders.
  3. Price a model switch. Compare cost per task across models for real workloads, including migration effort and output quality. Track whether cheaper alternatives suffice for specific jobs.
  4. Separate usage from returns. Track realized revenue or saved labor against serving and infrastructure costs. Treat usage growth and contracted demand as inputs, not proof of profitable utilization.

What to watch

  • Whether OpenAI or Anthropic runs into trouble within the next year, the window Eisman says matters most.
  • Whether Saffalo's end-2027 capacity estimate proves right and AI CapEx slows in 2028.
  • Whether model providers keep cutting prices as enterprise cost controls and cheaper alternatives spread.

Send this to your CFO and the leader budgeting AI infrastructure.

8:51

Watch the minute where Eisman gives the middle case a 40% chance

Written by Real Alpha from the episode's transcript and the other moments it links, then checked against the transcript. Quotes are the speaker's exact words.

Minute by minute

  1. 0:00AI demand depends on costly infrastructure

    Eisman says AI is influencing stocks, bonds and the US economy, but the investment boom depends on a few companies delivering. He explains LLMs, AGI and AI agents, citing his wife's use of an agent to find and buy theater tickets. He says a gigawatt-scale data center could cost around $30 billion and that hyperscalers are expected to spend roughly $850 billion on AI infrastructure in 2026.

  2. 5:49Eisman gives the intermediate case 40%

    Eisman assigns 30% each to the bull and bear cases, 40% to an intermediate case and about 0% to a doomsday scenario. In the intermediate case, data center construction lags or outpaces demand, Chinese models pressure prices, and AI spending slows; he cites an estimate that US data center capacity could be double demand by the end of 2027. In the bear case, weak returns could lead to recession, a market crash and losses for private credit and private equity.

  3. 9:57AI reaches most of the market

    Eisman estimates that AI-related companies account for well over 65% of the S&P 500, while AI-related bonds make up over half of this year's issuance. Bulls point to around $2 trillion in multi-year AI cloud-related backlog and argue that demand, rather than capital, is now running into power, permitting and memory constraints. Eisman compares the boom with the internet buildout, which transformed the world but also left many first-generation companies behind.

  4. 13:34Software companies face uneven AI risks

    Eisman says the SaaSpocalypse story—that AI will replace existing software—oversimplifies the risks: companies with high switching costs, such as ServiceNow and Salesforce, may be more protected. He notes that Adobe fell from $630 in February 2024 to around $245, and that Brandon Thomas released a free, open-source alternative to Creative Cloud. He argues that a 50% decline in software-company values could wipe out the equity in many private-equity deals, though he does not see this as large enough to seriously hurt the economy.

  5. 16:24AI makes hyperscalers capital-heavy

    Eisman says AI has shifted hyperscalers from cash-generating, mostly capital-light businesses to capital-heavy ones. He says a Blackwell GPU can cost more than $5,000 and a data center may need at least 100,000 of them; he also says Google raised $85 billion in equity in 2026 after previously generating over $80 billion in annual free cash flow. He notes that Oracle's debt rose from $94 billion to $155 billion in two years and that S&P downgraded it to BBB minus.

  6. 19:13Eisman questions financing structures

    Eisman says Meta is funding a $27 billion Louisiana data center through an off-balance-sheet vehicle: Meta owns 20%, while Blue Owl and Pimco own the rest, and Meta has committed to rent the facility for 20 years. He notes that Ernst & Young flagged the arrangement as a critical audit matter, and says Nvidia also invests in or lends to companies that then buy its chips. He argues these practices add risk but are not enough by themselves to sink the AI story.

  7. 22:27Power and permits could slow construction

    Eisman says permitting delays and local opposition, including concerns about water and energy use, could constrain data center construction. He notes that the electrical grid is already stressed and that building gas turbines takes years, while hyperscalers are also exploring nuclear power and alternatives such as Bloom Energy. He says he suspects construction is taking longer than hoped but cannot judge the overall impact.

  8. 24:25AI demand is concentrated

    Eisman says 70% of Nvidia's accounts receivable came from five customers, and cites Wall Street estimates that Anthropic and OpenAI generate about 70% of hyperscalers' AI revenue. He notes that OpenAI accounts for around half of Oracle's roughly $600 billion backlog and over 20% of CoreWeave's contracted order book. He argues that, because hyperscaler spending accounts for around half of projected US GDP growth in 2026, the economy depends heavily on those companies and their AI customers.

  9. 26:28Eisman questions OpenAI's resilience

    Eisman says OpenAI appears weaker than Anthropic and points to senior departures and reported June-quarter revenue of $6.7 billion, up 18% from March, versus Anthropic's revenue above $11 billion, up over 100%. He cites reported OpenAI costs of $12.3 billion, up $3 billion from March, and argues that its early focus on underpriced consumer products left it playing catch-up in business sales. He also relays Ed Zitron's estimate that OpenAI may need to raise $100 billion a year for several years, while saying it is too early to predict failure.

  10. 29:57Eisman warns OpenAI failure could hit markets

    Eisman says Oracle could be in immediate trouble if OpenAI fails, pointing to OpenAI's share of Oracle's backlog and Oracle's debt. He notes that Oracle's stock rose after its backlog surged in 2025, then gave back those gains after investors focused on its OpenAI exposure; he says it is now around 130, down roughly 70-ish percent from its peak. He argues that an OpenAI failure could prompt hyperscalers to cut spending and potentially push the US into recession, especially if it happens within a year.

  11. 32:05Investors could rotate toward safety

    Eisman says an AI downturn could hurt hyperscalers, banks and investment banks, and industrial companies tied to power, electrification and automation. He suggests investors might reallocate toward healthcare, consumer staples and safer subsectors such as property and casualty insurance. He names LVHD, SPLV and KBWP as examples of ETFs in those areas.

  12. 33:58Cheap models threaten AI pricing

    Eisman argues that AI models have low switching costs, so enterprise reliance may not be a lasting moat if competitors offer comparable quality at lower prices. He says open-weight models can cost around $0.20 per million tokens, compared with $2 to $10 for leading Anthropic and OpenAI models. He argues that this price gap, combined with limited moats, raises the risk of a price war that the model providers may not be able to withstand.

  13. 36:39Eisman favors infrastructure over model bets

    Eisman says he believes AI is real and will generate substantial profits, but the harder question is which companies will capture them. He says he is more interested in infrastructure suppliers such as Nvidia, Micron, Cisco, GE Vernova and Eaton, and more cautious about model providers and software businesses with low switching costs. He says stocks can disappoint even if the technology works, if reality falls short of expectations.

Published Nov 9, 2026.

Transcript · 28 min read

The Real Eisman Playbook, hosted by Steve Eisman, with Steve Eisman.

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

  1. AI has become the dominant theme in investing today. It's driving the stock market, influencing the bond market, and to a remarkable degree, driving the entire U.S. economy. But here's what concerns me. This entire AI investment boom has become increasingly dependent on the success of just a few companies. And if these companies don't deliver on their enormous promises, the consequences could be incredibly severe. To put it simply, the U.S. could easily enter a recession, and we could almost certainly see a large market correction. Hi, this is Steve Eisman, and welcome to another episode of The Real Eisman Playbook.

  2. Now, don't get me wrong. AI is an extraordinary technology, and I have no doubt it will change much of the world. But the questions investors need to be asking are how quickly, at what cost, and most importantly, who's actually going to make money from it? because right now, hundreds of billions of dollars are being poured into AI infrastructure, companies are taking on enormous amounts of debt, and investors are betting that all this spending will eventually generate extraordinary profits. So in this video, I want to examine exactly where we stand in the AI boom. We'll look at who's financing it all, the bull, bear, and doomsday scenarios, and most importantly, the five major risks that I believe could derail the entire investment story. Not since the railroad, as a new technology caused so much CapEx confusion, tension, and excitement. Since, for many, the whole AI story is rather overwhelming, I thought it would be a good idea to go through the major aspects of the story, and importantly, the major players. First things first, what exactly is an LLM? Well, LLM stands for Large Language Model.

  3. Now, I'm oversimplifying here, but think of it as an incredibly sophisticated word prediction machine. You ask it a question, and it predicts what word should come next to produce an answer. That's how ChatGPT and Claude work. That's different, the LLM, from AGI, or Artificial general intelligence. AGI is essentially the holy grail where machines could theoretically match or exceed human intelligence across a broad range of tasks. In the wonderful words of Spinal Tap, it goes to 11. numbers all go to 11. Now, we're not there yet. Although the people raising billions of dollars to develop it would certainly like you to believe we're getting close. Then there's agentic AI, which I think is particularly interesting. Whereas a traditional LLM answers your question, an AI agent can actually go out and perform tasks on your behalf. My wife, Valerie, for example, has used an AI agent to search for theater tickets, compare prices, and purchase the theater tickets she wanted with her permission.

Read the whole transcript
  1. That's genuinely useful technology. And I think it has enormous potential to change how we live and work. But here's where the investment story gets interesting. All of this requires extraordinary amounts of computing power. And that computing power lives inside data centers. Data centers are enormous facilities filled with sophisticated chips, servers, cooling equipment, and electrical infrastructure. And by the way, these things are incredibly expensive. A new large AI data center capable of supporting a gigawatt of power demand could cost around, get this, $30 billion to build and equip. That's just one data center.

  2. Now, you've probably heard that data centers are popping up across the country. and indeed, across the world. But who's actually paying for all of this? Primarily, not exclusively, but primarily, it's the hyperscalers. That's companies like Microsoft, Google, Amazon, and Oracle, which have spent years building the cloud computing infrastructure that businesses rely on. Except now, AI is forcing them to build data centers on an entirely different scale. In 2026 alone, hyperscalers, just the hyperscalers, are expected to spend roughly $850 billion on AI infrastructure and potentially even more next year.

  3. Astonishing sums. And that leads us to probably the two most important players in the current AI story. At the center of this entire AI ecosystem sit two private companies. OpenAI led by Sam Altman and Anthropic led by Dario Amodei. They're developing the AI models that are driving much of this demand for data centers. And possibly the kingpin of this whole shebang is Invis. NVIDIA creates the most sophisticated chips in the world, and Jensen Wang is at the epicenter of the AI build-out. So, you have the AI companies developing the models, the hyperscalers spending hundreds of billions building the infrastructure to support it, NVIDIA manufacturing billions of dollars in chips and financing them as well, and all kinds of debt and equity investors including private credit, banks, insurance companies, corporations who buy the debt issuances and stockholders all providing enormous amounts of capital to finance the whole thing.

  4. Which brings us to literally the trillion dollar question, who is actually going to make money from all of this? Because as we saw with the railroads and later the internet, a technology can completely transform the world while leaving an enormous trail of bankruptcies behind it. And right now, the market seems to have a few very different ideas about how this AI story will play out.

  5. Steve Eisman · 4 min

    “It was a good report where he estimated that U.S. data center capacity will be double the demand by the end of 2027.”

  6. I give this narrative about a 0% chance. Now that we've set the stage, let's turn to the AI investment story. Both the positive and the negatives. Over the past few years, AI has sent the stock market into a frenzy. In fact, from 2023 through 2025, the Magnificent Seven were responsible for roughly half of the S&P 500's total return. Hyperscalers are pouring hundreds of billions of dollars into AI data centers. At times, investors have become euphoric. But before we get to AI fundamentals, let me first

  7. Steve Eisman · 3 min

    “And add all that up, and you are well north of 65% of the S&P.”

  8. Steve Eisman · 3 min

    “Most transactions took place with 50% debt and 50% equity.”

  9. And that leads me to the risks. What are the major economic risks to the AI story? Well, there are five predominant issues. I'm going to examine them in depth and then rank that risk in terms of its ability to derail the AI story. Issue one, from capital light to raising capital. By tradition, the hyperscalers, Google, Microsoft, Amazon, Oracle, are mostly software companies. Yes, they created cloud businesses years ago that required building out data centers. But the cost of those data centers was not prohibitive.

  10. Steve Eisman · 6 min

    “In the past two years, Oracle's total debt has increased from $94 billion to $155 billion. The ratings agencies are not amused.”

  11. constraint is the lack of sufficient power and delays in building data centers. I needed to hammer this point home. The AI story cannot work unless AI data centers are built. Now the sums here are insane. Hyperscalers will spend $850 billion on building AI data centers just in 2026 and purchasing all the equipment to fill the data centers as well. But it takes time to build AI data centers. Here, politics plays a role. The AI industry is not popular. There is now a backlash against AI in many communities. People are afraid of the massive needs of AI data centers, needs for water and energy. So the process of permitting AI data centers is being delayed in many parts of the country. Energy is also a potential binding constraint. AI data centers utilize enormous amounts of energy. Now the first choice is to hook up the data center to the electrical grid, but the grid is already stressed. Utilities are investing in new plants, but that construction also takes time. It takes a few years, for example, to build a gas turbine that powers a

  12. utility plant. That is why a stock like GE Vernova has done well, because GEV is a large builder of gas turbines. Because utilities cannot satisfy all the needs of AI data centers, the hyperscalers are looking for alternatives. They are looking at nuclear or alternative gas turbine technology, such as those provided by Bloom Energy. My suspicion is that AI data center construction is taking longer than hoped, but how much of an impact this will have is something I just don't know. And now we turn to the third risk, concentration risk. Before we get to this

  13. Steve Eisman · 8 min

    “Hyperscalers are spending about $850 billion in CapEx this year, and even more next year. And that spending accounts for around half of the 2% GDP growth projected for 2026.”

  14. pull back on CapEx. Cloud revenue growth would slow. These stocks, Amazon, Google, Microsoft, and Oracle, just to name a few, they'd all correct. Also, the whole tech sector would correct, but the remember Gamifications are much broader than just that. Investment banks and large banks. These stocks are at fairly close to peak valuations. Also, the investment banking cycle is super strong, partially because of the financing needs of AI.

  15. Should those needs lessen, the investment banking cycle would weaken and these stocks would correct from their peak valuations. Industrials. There is a subset of industrial companies that are major beneficiaries of AI. They are in the power space, like GE for Nova and Quanta. Or they are in the electrification or automation spaces like Eaton and Rockwell. And these stocks would decline as well. So what will do well? This is not a stock picking question, but a reallocation question. Investors will reallocate to safety sectors and subsectors. In the safety sector category, think about health care and consumer staples.

  16. I'd also point out that within almost every sector, there exists safety subsectors. So, for example, within financials, the property and casualty sector is considered the safety subsector. Since we are talking about capital reallocation and not stock picking, let me flag three safety ETFs. One, LVHD, the Franklin U.S. Low Volatility High Dividend Index ETF. Number two, the SPLV, Invesco S&P 500 Low Volatility ETF. And last, KBWP, the Invesco KBW Property and Casualty Insurance ETF. And now we come to the last and probably the most important risk.

  17. Steve Eisman · 3 min

    “The flagship models of Anthropic and OpenAI can cost $2 to $10 per 1 million tokens. Openweight models can cost around $0.20 per 1 million tokens.”

  18. As you can see, there is a lot going on. The AI story is complex. It is also driving the U.S. economy right now. the risks I have outlined are real and could, I emphasize, could derail the AI story. If Anthropoc or OpenAI get into trouble in the next year, that would be telling. However, if it is years from now, the AI story may become so diversified by then that it won't matter. It is still too early to tell, but we will be watching very closely. So with that said, what should investors be doing right now? Well, the first thing I'd say is that being cautious about parts of the AI trade is not the same thing as being bearish on AI. I think AI is real. I think it's already started changing the economy, and I think an enormous amount of money will be made from it. The much harder question is, who actually makes that money?

  19. Personally, I'm much more interested in the picks and shovels. NVIDIA obviously sits at the center of the compute build-out. You've got companies like Micron supplying memory, Ariston, Cisco supplying networking. And then on the power side, you've got companies like GE Vernova and Eaton. These businesses don't necessarily need to predict which LLM ultimately wins. They're selling the infrastructure required to build the entire thing. Where I'd be much more careful is in businesses where investors are assuming enormous future profits despite very low switching costs. Who are we talking about here?

  20. Mainly the LLM providers themselves, but potentially to parts of software too as agentic AI becomes more capable. The question I keep coming back to personally when analyzing these businesses is very simple. How difficult is it for the customer to leave? If the answer is not very, you'd better be sure about the price you're paying. And ultimately, that's how I'm looking at this entire AI boom. I'm certainly not trying to call it top. I'm not telling you this is 1999 or 2008 all over again, at least not yet anyway. Maybe these companies deliver everything the market is expecting and more. But when valuations assume an extraordinary future, you don't necessarily need the technology to fail for the stocks to disappoint.

  21. You just need reality to fall short of expectations. And right now, between Chinese competition, the lack of moats, the enormous commitments sitting on OpenAI and anthropic, and the physical limitations of the power grid, there are a plethora of reasons to keep watching those expectations very carefully. And that's the end of our lecture. But before I go, let me just say that if you want to support the show further and get additional exclusive content, you can check out my sub stack at steveiseman.com. Premium membership to Substack offers a premium-only Wednesday episode, which can be a masterclass or deep-dive interview, and premium subscribers receive early access to our Monday interviews and Friday weekly wraps, as well as full access to my daily chats and the ongoing conversations among members.

  22. Free membership in Substack offers access to my daily notes and previews of our Wednesday premium interview and previews of all newsletters. The link, steveiseman.com, is in the description. Thanks for watching. This podcast is for informational purposes only and does not constitute investment advice. A host and guests may hold positions in stocks discussed. Opinions expressed on their own and not recommendations. Please do your own due diligence and consult a licensed financial advisor before making any investment decisions.

More interviews