Key Moments
Did the AI Bubble Just Pop?! Anthropic's Leaked Numbers are INSANE
Impact TheoryKey Moments
AI's monumental debt-fueled expansion is poised for a dramatic correction, with leaked numbers revealing Anthropic's $42 billion loss on $4.6 billion revenue and $518 billion in future commitments, indicating a potential financial collapse.
Key Insights
Anthropic's leaked S-1 filing reveals a staggering $42 billion loss in 2025 against $4.6 billion in revenue, with an additional $518 billion committed to future spending over the next two years.
The AI bubble is structured as a debt story, where companies are taking on unprecedented levels of debt to fund revolutionary technology, mirroring historical bubbles like the dot-com crash.
The dot-com era's failures offer a cautionary tale, with companies like Nortel and Lucent providing direct loans to customers, a mechanism now echoed in AI through mechanisms like residual value guarantees.
The current AI boom is characterized by a massive capital inflow, with investors pouring money into AI companies, driven by the belief in a transformative technological shift and the potential for rapid intelligence generation.
The financial complexity, including the use of Special Purpose Vehicles (SPVs) and private credit markets, makes the AI bubble hard to regulate and understand, increasing systemic risk.
The sustainability of profits in AI is questioned, with long-term value likely residing in the infrastructure layer rather than individual AI models if they become commoditized and open-source models proliferate.
The unsustainable debt fueling the AI expansion
The current AI boom is fundamentally a debt story, characterized by companies taking on unprecedented levels of debt to fund revolutionary technology. This mirrors historical bubbles, with the dot-com crash serving as a stark warning. The core issue is the massive capital inflow into AI, driven by the conviction that this technology will fundamentally change the world and enable rapid intelligence generation. However, this expansion is being financed through substantial debt, creating a precarious situation where the ability to repay this debt is contingent on projected revenue growth that may not materialize. The economic laws of physics, though complex and interacting with human behavior, are described as mechanical, and ignoring these mechanisms, particularly the accumulation of debt, is extremely risky. Two primary outcomes for unpayable debt are a sharp default leading to system collapse or inflation as a soft default, both with significant consequences.
Anthropic's leaked financials reveal the scale of the crisis
Leaked S-1 filings from Anthropic provide a concrete look at the financial realities underlying the AI boom. The company reported revenues of approximately $4.6 billion in 2025, a twelvefold increase, which is undeniably impressive. However, this was accompanied by operating losses of $8 billion. When accounting for capital expenditures and other expenses, the total loss for 2025 ballooned to $42 billion. A significant portion of this $42 billion loss is considered a non-cash, paper loss related to the valuation of their stock and potential future repurchase obligations. The more tangible figure to focus on is the $8 billion cash burn against $4.6 billion in revenue. Furthermore, Anthropic has committed to spending $518 billion over the next two years. This colossal commitment, relative to its current revenue, highlights the extreme capital expenditure required to scale AI technologies and raises serious concerns about the sustainability of such an expansion.
Historical parallels: The dot-com bubble's cautionary tale
The current AI bubble shares striking similarities with the dot-com era, particularly in its debt-driven structure and reliance on future revenue projections. During the dot-com boom, companies like Nortel and Lucent engaged in practices like offering direct loans to their customers to facilitate equipment purchases. This is now mirrored in the AI space through mechanisms such as residual value guarantees, where established companies like NVIDIA and Broadcom provide support for loans taken out by AI firms. These guarantees aim to make debt cheaper for AI companies by backstopping the value of the purchased equipment. However, just as the dot-com bubble burst when the projected revenue growth failed to materialize and the debt became unpayable, the AI bubble faces a similar risk. The complexity of these financial arrangements, involving Special Purpose Vehicles (SPVs) and private credit markets, obscures the true risk and makes regulation difficult, echoing the systemic vulnerabilities of the past.
The role of debt and credit markets in the AI expansion
The AI expansion is heavily reliant on debt and credit markets. Companies are aggressively seeking funding, with Anthropic alone reportedly valued at close to $1 trillion in its last private funding round and aiming for a $2 trillion market cap. This massive valuation is supported by future revenue projections, not current profitability. The sheer scale of committed spending, $518 billion for Anthropic, necessitates significant debt financing. If revenue growth falters, these companies will struggle to service their debt, potentially leading to a liquidity crisis. The complexity of the financial structures, including SPVs and private credit, allows for risks to be hidden and amplified. As interest rates rise, the cost of this debt increases, further pressuring companies and potentially leading to a pullback from lenders. The interconnectedness of these financial players means that a failure in one area can have cascading effects across the system.
The 'this time is different' fallacy and the inevitability of correction
A recurring theme in speculative bubbles is the belief that 'this time is different.' While AI is a revolutionary technology, the economic mechanisms at play are not entirely new. The AI bubble is characterized by an intense competition for capital, with a 'black hole' effect drawing in global liquidity. Investors are convinced of a rapid and transformative shift, leading them to act, in some views, recklessly. However, historical patterns suggest that such rapid, debt-fueled expansions are susceptible to sharp corrections. The argument is that even with groundbreaking technology, the laws of economics will eventually assert themselves. A potential AI collapse is viewed by some as a necessary event to clear out excesses and allow the market to find a more sustainable path, though a slow deflation of the bubble is also considered.
Potential fallout and the question of sustainable profit
The implications of a potential AI bubble pop are immense, with trillions of dollars in investor capital at risk. The market is pricing in perfection, with little room for error. If AI models become commoditized and open-source alternatives gain traction, sustainable profits may shift from AI developers to the infrastructure layer. Companies that can offer the best products at the best prices, or those providing essential infrastructure like cloud services and data centers, are seen as potentially safer bets. However, even infrastructure investments carry risks, especially if they are tied to specific geographic locations or face unforeseen challenges. The core issue remains the immense debt burden and the uncertainty surrounding revenue projections, making the entire sector vulnerable to shifts in market sentiment and credit availability.
Regulatory capture and political implications
The AI industry is also navigating a complex political landscape, with companies like Anthropic and OpenAI potentially engaging in 'regulatory capture.' The debate over AI regulation could become a political battleground, with different parties advocating for distinct approaches. One side might push for stricter government oversight to control risks, while another might favor a more laissez-faire approach to maintain U.S. competitiveness, particularly against China. This political dynamic could significantly impact the distribution of value and the competitive landscape of the AI industry, influencing which companies succeed and how the technology is deployed. The outcome of these political debates could determine whether AI development leads to widespread prosperity or concentrated power.
Mentioned in This Episode
●Products
●Companies
●Organizations
●Books
●Concepts
●People Referenced
Anthropic Financial Snapshot (2025 Estimates)
Data extracted from this episode
| Metric | Value | Notes |
|---|---|---|
| Revenue | 4.6 Billion USD | 12x increase from previous year |
| Operational Losses | 8 Billion USD | Actual cash outflow |
| Total Losses (including CAPEX) | 42 Billion USD | Primarily non-cash/paper losses |
| Committed Spending (Next 2 Years) | 518 Billion USD | Half a trillion dollars |
Common Questions
The discussion suggests a high probability of an AI bubble bursting, drawing parallels to the dot-com crash. The rapid accumulation of debt and the financial metrics of companies like Anthropic indicate unsustainable growth, leading to predictions of a significant collapse.
Topics
Mentioned in this video
Mentioned as someone who agrees with the sentiment that the AI bubble's collapse, the sooner the better, will prevent further destruction of investor capital.
Mentioned in the context of inflation and inelastic demand, suggesting a potential driver for continued debt accumulation.
His response to questions about job displacement is characterized as steering the conversation towards an 'age of abundance' rather than directly addressing concerns about employment.
Reported on the leaked S-1 filing of Anthropic, providing insights into the company's financial situation.
Its need to raise interest rates to attract buyers for US debt is seen as competing with the AI industry's borrowing needs, exacerbating inflationary pressures.
A firm in the private credit market that has reportedly shown errors in its records, indicating potential issues with loan defaults.
Its actions to raise interest rates to combat inflation are presented as being in conflict with the AI industry's continued borrowing and spending, creating an 'inflationary arms race'.
Cited as an example of a stable company with sustainable revenue streams, contrasting with highly speculative AI investments.
Mentioned as a company involved in financing structures for AI companies, providing residual value guarantees for debts.
Mentioned as a participant in private credit deals related to AI companies.
A prominent company in the AI bubble, whose leaked S-1 filing revealed significant financial data, including high revenues but also massive operational and capital expenditures leading to substantial losses.
Mentioned as an arranger of private credit deals, including those involving AI companies.
Its credit rating was downgraded, serving as a cautionary example of a company that over-leveraged with debt.
Mentioned alongside Anthropic as playing the 'regulatory capture' game, influencing political discourse around AI regulation.
Mentioned as a company providing chips that are essential for AI development and potentially supporting financing structures through guarantees.
Cited as an example of a stable company with sustainable revenue streams, contrasting with highly speculative AI investments.
A historical example from the dot-com era, similar to current AI companies in its business model of directly financing customers and providing loans for equipment purchases.
Its collapse in 2008 is used as an example of how a sudden loss of confidence in the debt market can trigger a financial crisis.
Its CEO reportedly stated in December that the numbers for AI do not make sense, reflecting skepticism about the financial projections.
A historical example from the dot-com era, similar to current AI companies in its business model of directly financing customers and providing loans for equipment purchases.
The speaker's former company, which experienced rapid growth (57,000% in three years), illustrating that extraordinary growth is possible but carries significant risk and requires careful financial management.
A firm in the private credit market that has reportedly shown errors in its records, indicating potential issues with loan defaults.
Its CEO's reaction after the dot-com bubble is referenced to highlight the unrealistic expectations placed on businesses, with current AI numbers being even more extreme.
Mentioned as a potential customer for Broadcom chips, fitting into the complex financing structures described.
Entities used in complex financial structures, particularly in private credit markets, to pool investments and manage debt, mirroring structures seen in the 2008 mortgage crisis and dot-com bubble.
Mentioned as the goal of AI development, with the potential to create 'thousands, tens of thousands, hundreds of thousands of smart copies' instantly, driving massive demand for computing power.
Mentioned as a potential societal response to widespread job displacement caused by AI and automation.
A future scenario envisioned by figures like Elon Musk, characterized by universal basic income and high living standards, proposed as a potential outcome of advanced AI and automation.
Discussed in the context of China's strategy to use them to undermine revenue models of AI companies and create a competitive advantage.
A financial mechanism where a company guarantees the value of assets used as collateral, thereby making debt cheaper for the borrower and providing a backstop for lenders.
More from Tom Bilyeu
View all 169 summaries
105 minTHIS is What Happens When AI Gets Smarter Than Humans
47 minThe Japan Playbook Is Coming to America — The Big Reset Is Here
116 minIran’s Problems are Worse Than You Think & OpenAI Had to Hit the Panic Button AGAIN
131 min10 YR Treasury Yield is Going Vertical, Speedrunning 2008 Financial Crisis
Ask anything from this episode.
Save it, chat with it, and connect it to Claude or ChatGPT. Get cited answers from the actual content — and build your own knowledge base of every podcast and video you care about.
Get Started Free