Bloomberg reports that OpenAI expects to reach at least $70 billion in annualized revenue by the end of 2026, up from roughly $50 billion at the end of September, with enterprise sales driving the expected growth. The crucial distinction is that $50 billion is a reported current run rate, while $70 billion is a year-end forecast—not revenue already earned in 2026. 12

• Why the figures differed: Axios says the roughly $70 billion September figure it and others reported earlier came from information shared with investors using a calculation intended to make OpenAI more comparable with Anthropic. The Financial Times’ later roughly $50 billion figure reflected OpenAI’s reported revenue run rate; Bloomberg subsequently reported $70 billion as a future year-end target. Those are different measures and dates, not evidence that $20 billion of sales disappeared. 1210
• The accounting comparison: As described in the reports, OpenAI records certain sales through Microsoft net of the partner’s share, while Anthropic records some sales through Amazon and other cloud partners gross. Gross reporting makes the headline sales figure larger without necessarily changing the underlying customer spending; both approaches are reported to comply with U.S. accounting standards. Anthropic’s reported $65 billion July run rate therefore should not be treated as directly comparable with OpenAI’s net September figure without reconciling the partner-sales treatment. 10
• Growth and markets: Reports put OpenAI’s third-quarter run-rate growth at 77% overall and 107% for enterprise, though those growth claims do not establish that the year-end target will be met. The lower September figure was treated as a warning for the AI-spending trade: Thursday’s reported declines were 1.4% for the Nasdaq 100 and 3.4% for chip stocks, with Oracle, Intel and Nvidia among those hit. Bloomberg’s subsequent year-end forecast eased that concern, and reports described a rebound in Nasdaq futures and in Nvidia, Micron and other technology shares. These are reported market reactions, not proof that one disclosure alone caused every move. 178
• Funding and the unresolved risk: Reports describe investor disclosures behind the revenue figures, but OpenAI has not publicly confirmed the detailed accounts and declined to comment. It is reportedly seeking at least $30 billion at a $1.4 trillion pre-money valuation, potentially with UAE investors including MGX, after a reported $122 billion March raise at an $852 billion post-money valuation. Altman has ruled out a 2026 OpenAI IPO, while an Anthropic listing as early as November remains a possibility, not a commitment. The central unanswered question is whether rapidly growing revenue and new financing can cover compute costs and continuing losses; the cited $280 billion cash-burn-through-2030 figure is a projection, not an observed outcome. Insufficient evidence to conclude that the spending path is sustainable. 12

Bloomberg reports that OpenAI expects to reach at least $70 billion in annualized revenue by the end of 2026, up from roughly $50 billion at the end of September, with enterprise sales driving the expected growth. The crucial distinction is that $50 billion is a reported current run rate, while $70 billion is a year-end forecast—not revenue already earned in 2026. 12

• Why the figures differed: Axios says the roughly $70 billion September figure it and others reported earlier came from information shared with investors using a calculation intended to make OpenAI more comparable with Anthropic. The Financial Times’ later roughly $50 billion figure reflected OpenAI’s reported revenue run rate; Bloomberg subsequently reported $70 billion as a future year-end target. Those are different measures and dates, not evidence that $20 billion of sales disappeared. 1210
• The accounting comparison: As described in the reports, OpenAI records certain sales through Microsoft net of the partner’s share, while Anthropic records some sales through Amazon and other cloud partners gross. Gross reporting makes the headline sales figure larger without necessarily changing the underlying customer spending; both approaches are reported to comply with U.S. accounting standards. Anthropic’s reported $65 billion July run rate therefore should not be treated as directly comparable with OpenAI’s net September figure without reconciling the partner-sales treatment. 10
• Growth and markets: Reports put OpenAI’s third-quarter run-rate growth at 77% overall and 107% for enterprise, though those growth claims do not establish that the year-end target will be met. The lower September figure was treated as a warning for the AI-spending trade: Thursday’s reported declines were 1.4% for the Nasdaq 100 and 3.4% for chip stocks, with Oracle, Intel and Nvidia among those hit. Bloomberg’s subsequent year-end forecast eased that concern, and reports described a rebound in Nasdaq futures and in Nvidia, Micron and other technology shares. These are reported market reactions, not proof that one disclosure alone caused every move. 178
• Funding and the unresolved risk: Reports describe investor disclosures behind the revenue figures, but OpenAI has not publicly confirmed the detailed accounts and declined to comment. It is reportedly seeking at least $30 billion at a $1.4 trillion pre-money valuation, potentially with UAE investors including MGX, after a reported $122 billion March raise at an $852 billion post-money valuation. Altman has ruled out a 2026 OpenAI IPO, while an Anthropic listing as early as November remains a possibility, not a commitment. The central unanswered question is whether rapidly growing revenue and new financing can cover compute costs and continuing losses; the cited $280 billion cash-burn-through-2030 figure is a projection, not an observed outcome. Insufficient evidence to conclude that the spending path is sustainable. 12