Companies like Google and Microsoft face a dilemma: use their compute to develop their own AI models or rent it out for high returns. The profitable infrastructure-as-a-service model often wins, starving internal research teams and creating a conflict that slows their model development, an issue pure-play labs like OpenAI don't face.

OpenAI and Anthropic are presenting a version of profitability that excludes their largest expenses: model training and inference. Critics compare this to an airline ignoring the cost of its jets. This financial engineering aims to create a positive outlook for potential IPOs but masks their true cash burn rate.

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.

Contrary to the "bubble pop" narrative, a market shift away from high-margin frontier models toward cheaper alternatives could boost overall AI usage. This would redirect revenue from labs like OpenAI to infrastructure players who provide the most efficient, low-cost compute.

An AI lab's P&L contains two distinct businesses. The first is training models—a high upfront investment creating a depreciating asset. The second is the 'inference factory,' a profitable manufacturing business with positive margins. This duality explains their massive losses despite high revenue.

The paradoxical financial state of AI labs: individual models can generate healthy gross margins from inference, but the parent company operates at a loss. This is due to the massive, exponentially increasing R&D costs required to train the next, more powerful model.

Software has long commanded premium valuations due to near-zero marginal distribution costs. AI breaks this model. The significant, variable cost of inference means expenses scale with usage, fundamentally altering software's economic profile and forcing valuations down toward those of traditional industries.

AI companies like OpenAI are losing money on their popular subscription plans. The computational cost (inference) to serve a user, especially a power user, often exceeds the subscription fee. This subsidized model is propped up by venture capital and is not sustainable long-term.

While profitable on their last model, AI companies are "borrowing against the future." The cost of training their next-generation models makes them currently unprofitable. Their business model relies on perpetually raising larger rounds, a dependency that creates systemic market risk.