Contrary to the narrative of burning cash, major AI labs are likely highly profitable on the marginal cost of inference. Their massive reported losses stem from huge capital expenditures on training runs and R&D. This financial structure is more akin to an industrial manufacturer than a traditional software company, with high upfront costs and profitable unit economics.

Traditional accounting metrics misrepresent the financial health of AI companies. Their largest expenditure, acquiring compute power, should be viewed as an investment in a valuable, appreciating asset, not as a typical operating expense. This reframes the narrative around their massive cash burn.

A massive portion of cloud providers' growth comes from just two AI companies, OpenAI and Anthropic. Since these same providers (e.g., Microsoft, Google) are also major investors in those startups, it creates a circular economy where investment capital flows directly back as revenue for compute.

As long as every dollar spent on compute generates a dollar or more in top-line revenue, it is rational for AI companies to raise and spend limitlessly. This turns capital into a direct and predictable engine for growth, unlike traditional business models.

A new generation of AI-native companies is fundamentally restructuring its cost base. Instead of hiring more knowledge workers, they are allocating significant portions of their budget—up to 30% of what would be spent on compensation—directly to AI token consumption, driving massive productivity gains.

For decades, you couldn't catch a competitor with a two-year lead just by hiring more engineers. AI changes this. Access to massive capital for compute (GPUs) and data now allows teams to solve problems and close gaps quickly, making capital itself a primary competitive moat.

While a megawatt of compute costs ~$15M, leading labs like Anthropic can generate up to $50M in revenue from it. This massive 3x+ profit margin creates a powerful flywheel, allowing them to reinvest heavily in training the next generation of models and accelerate their lead.

The massive growth in AI token consumption isn't a sign of waste but of ambition. While the cost per "unit of intelligence" is decreasing, companies are immediately applying that efficiency to solve exponentially harder problems. Our appetite for more capable AI is growing faster than the cost is falling, leading to sustained, exponential spending.

For the first time, investors can trace a direct line from dollars to outcomes. Capital invested in compute predictably enhances model capabilities due to scaling laws. This creates a powerful feedback loop where improved capabilities drive demand, justifying further investment.