Did the AI Bubble Just Pop?! Anthropic's Leaked Numbers are INSANE

The "This Time Is Different" Fallacy [00:00:00]

• The common refrain that "this time is different" is often heard during speculative booms, but economic "physics" remain constant.
• Human sentiment complicates economic tracking, but the underlying mechanisms are mechanistic and should not be ignored.
• Debt accumulation is a predictable problem, with hard defaults and inflation as the primary ways to address it.

  "You lose sight of the mechanisms of the economy at your peril."

The AI Capital Hoover [00:02:00]

• AI is absorbing an unprecedented amount of global capital, creating a "black hole" effect.
• Companies taking on extraordinary debt for new technologies historically face brutal consequences.
• The longer the AI boom continues, the more devastating a potential bust will be.

  "AI is hoovering up all the capital in the world and it's subjecting itself to the mechanisms of debt."

Biblical Capital Destruction [00:03:00]

• The amount of investor capital that could be destroyed by an AI bust is described as "biblical."
• While AI will be a game-changing technology, the path to it involves significant financial destruction.
• Not all companies will win, and many investments will be lost, a pattern consistent with past bubbles.

  "The amount of capital from investors that will get destroyed is biblical."

The Anthropic Leak [00:05:00]

• Reuters reported on Anthropic's leaked S1 filing, providing crucial financial insights.
• Private companies typically disclose financials only after going public via IPO.
• The leaked data offers a glimpse into the financial realities of a major AI player.

  "Reuters basically gave us some insight into one of the biggest names in AI."

Staggering Financials [00:06:00]

• Anthropic's S1 reveals staggering numbers, including $518 billion in promised spending over the next few years.
• The company's revenue grew 12-fold to nearly $4.6 billion in the last year.
• Despite impressive revenue growth, the scale of committed spending raises serious concerns.

  "The numbers are just absolutely staggering."

Astronomical Losses [00:07:00]

• Anthropic incurred operating losses of $8 billion on $4.6 billion in revenue.
• Expenses totaled $12 billion against $4.6 billion in revenue.
• The overall loss, including capital expenditures, reached $42 billion in 2025.

  "So if you have to spend whatever, 12 to make five, I think it's a little bit different than that, but that's direction incorrect."

Paper Losses vs. Real Losses [00:08:00]

• A significant portion of Anthropic's $42 billion loss is a "non-cash paper loss" related to stock valuation and potential share buybacks.
• Operating losses of $8 billion represent real cash deficits leaving the business.
• The distinction is crucial for understanding the company's immediate financial health.

  "This is not actually dollars and cents that are leaving their bank account now."

Half a Trillion Committed [00:10:00]

• Anthropic has committed $518 billion in spending over the next couple of years.
• This commitment is against $4.6 billion in revenue from the previous year.
• The market's perception of AI as a "once in a civilization moment" fuels this massive investment.

  "These people have $4.6 billion in revenue last year and they've already committed half a trillion dollars."

The Reckless Bet on AI [00:12:00]

• The scale of the bet on AI is immense, with investors being "reckless" due to conviction in its transformative power.
• The race to develop AI is fueled by the need for compute power, data centers, and capital.
• The $518 billion figure highlights the enormous financial commitments being made.

  "And everybody is so convinced that this transition is going to happen on a reasonable timeline that they are being really, from where I'm sitting, reckless."

Valuation vs. Reality [00:13:00]

• Anthropic is seeking a $2 trillion market value, despite $5 billion in revenue and $40 billion in losses.
• The company's valuation is "priced to perfection," with no room for error.
• A disclosed risk in their S1 filing is the potential for their technology to "end humanity."

  "It is price to ultra perfection. There's absolutely nothing can go wrong."

The Dot-Com Pain [00:15:00]

• Wishing for an AI bust "sooner rather than later" could lead to 20 years of economic pain, similar to the dot-com bust.
• A slow deflation of the bubble might be preferable to a sudden crash.
• The risk of ushering in uncontrollable AI (Skynet) is a concern for some.

  "If you go back and look at what happened during the dot-com bust, it ended up being 20 years of pain."

Political Collision Course [00:18:00]

• Major AI players like Anthropic and OpenAI are engaging in "regulatory capture" games.
• This political maneuvering could pit Democrats against Republicans, influencing future elections and legislation.
• Democrats may argue for control and regulation, while Republicans might prioritize competition with China and market access.

  "The Anthropic and OpenAI are playing a regulatory capture game that is going to pit the Democrats against the Republicans."

The Infrastructure Play [00:24:00]

• If AI models become commoditized, durable profits may accrue to the infrastructure layer (e.g., cloud providers, energy companies).
• Companies that provide essential infrastructure are seen as a safer bet, though potentially with less upside.
• The geographic distribution of infrastructure, including potential expansion into space, adds another layer of uncertainty.

  "The infrastructure certainly feels like it is likely to be a safer bet."

The Debt Story, Amplified [00:26:00]

• The AI boom is fundamentally a "debt story," significantly larger and more reliant on the credit system than the dot-com era.
• Special Purpose Vehicles (SPVs) and complex guarantees are central to this financing structure, mirroring the mortgage bubble.
• Companies like NVIDIA and Broadcom are providing backstops and guarantees, deepening the entanglement.

  "This one is much, much bigger debt, much, much bigger reliance on the credit system as a whole."

Circular Financing and Entanglement [00:28:00]

• Deals involve companies lending to each other, creating circular financing where wealth creation is questionable.
• The entanglement is so deep that tracking who owes what is becoming incredibly complex, potentially requiring AI itself.
• Guarantees and backstops from major players like Broadcom are used to "cheapen" debt for AI companies.

  "So if you’ve got, let's say, NVIDIA lending money to Anthropic, knowing that Anthropic is going to take the money that they just lent them and buy their chips, now you just have NVIDIA gave money to themselves."

The Dot-Com Comparison [00:32:00]

• The structure of AI financing, particularly the use of SPVs and customer financing, echoes the dot-com era.
• Companies like Nortel and Lucent provided direct loans to customers to buy their products, a model now seen with NVIDIA and Broadcom.
• The key difference is the sheer scale of debt and complexity in the current AI boom.

  "The only real true difference, which is, in this case, from a macroscopic perspective, isn't a meaningful difference, is that these companies, the modern version of Lucid and Nortel, in this case, NVIDIA or Broadcom, you can substitute Broadcom in there, they're not making these loans directly."

Priced to Perfection, Not Realism [00:35:00]

• The AI boom is "priced to perfection," meaning it assumes everything will go exactly right.
• This unrealistic pricing, combined with massive debt, creates a high risk of failure.
• Even unprecedented growth doesn't guarantee success if debt servicing outpaces revenue.

  "It is a, I think, far better way to say it is it's not realistically priced."

The Credit Cycle's Turn [00:38:00]

• A loss of faith in the debt market, similar to what happened with Lehman Brothers, could trigger a collapse.
• If lenders become unwilling to provide debt at the required rates or amounts, companies can fail.
• The interconnectedness means a problem in one area can cascade through the entire system.

  "And when that happens, when you don't have the level of profitability that you need and the level of profitability that you need just continues to grow and grow and grow because of the amount of debt that you brought on, the amount of service that you have to do to that debt, now you really run into a problem."

The Uncertainty of Revenue [00:40:00]

• A major risk is the unpredictable nature of future revenue for AI companies.
• Factors like macroeconomic downturns, competition from lower-tier models, and interest rate hikes add to the uncertainty.
• The current environment sees a potential conflict between the Fed's efforts to curb inflation and the AI industry's demand for borrowing.

  "And the revenue has to come as projected and as extrapolated, which is another problem because nobody has any frigging clue what revenue is going to look like."

The Risk of Systemic Failure [00:43:00]

• The complexity and scale of the AI financing ecosystem make it difficult to regulate and prone to systemic failure.
• The "house of cards" structure means that a slowdown in money flow or a loss of confidence can lead to a rapid collapse.
• The historical precedent of the dot-com bust, where optimistic extrapolations failed to materialize, serves as a stark warning.

  "It's so big and so complex that, yeah, this is how this stuff becomes a systemic, a risk of systemic failure."

The Job Market Question [00:48:00]

• A significant concern is the impact of AI on the job market, particularly white-collar jobs.
• While AI promises an "age of abundance," the immediate question for many is their future employment.
• The potential for widespread job displacement raises questions about economic and social structures like Universal Basic Income (UBI).

  "No, motherfucker. It wasn't a question about the age of abundance. It was a question about what am I going to do? I've got a job. It's me on a keyboard. What am I going to do with my life?"

Navigating the AI Transformation [00:52:00]

• Investors should approach AI with caution, understanding the historical precedents of boom and bust cycles.
• While AI is a transformative technology, the path to its success is fraught with risk.
• Having "dry powder" to invest during a crash, while maintaining belief in the long-term thesis, was key in past booms.

  "But man, history has a brutal way of reminding people that these are mechanisms."

Where Does Durable Profit Accrue? [00:54:00]

• If AI models become interchangeable, durable profits are likely to accrue to the infrastructure layer.
• Companies providing essential services like electricity and cloud computing may be safer bets.
• The potential for AI to move into space adds a layer of complexity to infrastructure investment.

  "So the infrastructure certainly feels like it is likely to be a safer bet."

The 2008 Parallel [00:56:00]

• The 2008 financial crisis was triggered by a loss of faith in the debt market, leading to a credit freeze.
• The current AI boom's massive debt obligations, potentially reaching trillions, could face a similar fate if confidence erodes.
• The uncertainty surrounding these obligations makes the situation precarious.

  "When you look back at what ended up being the disaster of the 2008 great financial crisis, it was just people lost faith and once you lose faith, you don't make the debt available."