Documents: OpenAI's projections suggest it won't turn a profit until 2029, when its revenue would hit $100B; losses could hit $14B in 2026, up nearly 3x on 2024
Cory Weinberg / The Information :
Context & Ripple Effects
OpenAI had previously told investors it expected revenue to rise from $200 million in 2023 to $1 billion in 2024, making these documents a sharper test of how quickly sales must scale relative to spending. Those early revenue targets framed growth; the new projections frame the cost of reaching durable profitability.
The longer arc in the related coverage is one of rising targets alongside a heavier cost base: a later plan put 2030 revenue at $200 billion while assigning roughly 45% of it to R&D. That later R&D outlook suggests this is not simply a near-term loss cycle but a business model being built around sustained compute and research investment.
First-order effects
- OpenAI's financing case becomes immediately more dependent on investors' willingness to fund years of projected losses before a 2029 profit inflection.
- The projected 2026 loss profile raises the revenue and cost-control threshold OpenAI must meet to validate its path to $100 billion in revenue.
Second-order effects
- Investors and commercial partners gain a clearer benchmark for judging frontier-AI economics, likely focusing more closely on whether revenue growth can outrun training and serving costs.
- The projections put added pressure on rival AI developers to distinguish their own route to breakeven; later documents, for example, showed differing projected breakeven timelines for OpenAI and Anthropic. Those competing profitability forecasts make capital efficiency a comparative issue.
Third-order effects
- If such long loss periods persist across frontier models, access to patient capital and infrastructure financing could become as decisive as model quality in determining which providers remain independent.
- The sector may increasingly separate reported operating performance from the economics of training and inference, particularly as later disclosures showed inference costs exceeding half of revenue. That cost reporting distinction could become central to how investors evaluate AI businesses.
The trend: Frontier AI is evolving into a capital-intensive infrastructure business in which very large revenue ambitions are paired with delayed profitability and sustained compute costs.