- More than 50 percent of ChatGPT‘s regular users now have female first names, reversing an 80-20 male split at the platform’s late-2022 launch.
- The demographic shift has been in place since at least fall 2025, with OpenAI estimating nearly half a billion women use ChatGPT weekly.
- A separate OpenAI analysis estimates China’s total AI spending for 2025 at between $97.2 billion and $125.3 billion.
- OpenAI’s own compute capacity grew from 0.2 gigawatts in 2023 to approximately 1.9 gigawatts in 2025, with a stated 30-gigawatt target by 2030.
What Happened
OpenAI disclosed that more than 50 percent of ChatGPT’s regular users now have female first names, according to the company’s own data reported by The Decoder on April 17, 2026. The reversal has been in place since at least fall 2025. When ChatGPT launched in late 2022, roughly 80 percent of its users had typically male first names.
With weekly active users approaching one billion, OpenAI estimates that nearly half a billion women now use the platform regularly — a user count the company frames as larger than many national populations.
Why It Matters
OpenAI attributes the demographic shift to ChatGPT’s evolution from a technical niche product to an everyday tool, as generative AI became more socially familiar and early adoption barriers fell. The company also warns against assuming remaining demographic gaps — across income, education, company size, sector, and geography — will close without active monitoring and intervention.
The disclosure comes as consumer AI platforms face increasing scrutiny over who their products serve and how. A majority-female user base carries direct implications for product design, content policy, and safety system calibration.
Technical Details
The same report includes OpenAI’s estimate of China’s total AI spending for 2025: between $97.2 billion and $125.3 billion, drawn from corporate, direct government, and state-owned enterprise sources. Chinese corporate spending — led by Alibaba, ByteDance, Huawei, and Tencent — accounts for $67.5 billion to $76 billion of that figure. OpenAI notes it counted only direct government subsidies, meaning actual state spending is likely higher than the reported $16.8 billion to $36.4 billion range.
The US leads in capital expenditure with a 2026 forecast of $527 billion, though OpenAI acknowledges China’s lower infrastructure costs provide greater purchasing power per dollar. For 2026, OpenAI projects Chinese corporate AI spending will grow 17 percent, reaching between $78.3 billion and $89.4 billion.
On compute, OpenAI reports its available capacity grew from 0.2 gigawatts in 2023, to 0.6 gigawatts in 2024, to approximately 1.9 gigawatts in 2025. The company signed partnerships with Nvidia, Broadcom, Oracle, and AMD over the course of 2025, citing a combined pipeline of over 30 gigawatts on paper, with more than 8 gigawatts described as already “identified” toward a 2030 target.
Who’s Affected
Consumer AI developers and enterprise platform teams will need to account for a majority-female user base when designing interfaces, evaluating harms, and setting content moderation thresholds. The demographic data also affects how advertisers, researchers, and regulators frame their analysis of large-scale AI adoption.
On the compute buildout, Anthropic CEO Dario Amodei has publicly characterized OpenAI’s rapid infrastructure expansion as high-risk, stating the company is “doing stuff because it sounds cool.” Anthropic has said it is planning a more measured buildout of at least 10 gigawatts over the next several years. Reports also indicate OpenAI is pulling back on planned compute purchases in the European Union.
What’s Next
OpenAI has not announced a specific timeline for releasing further demographic breakdowns or methodology details behind its user-name analysis. The company’s stated goal of reaching one billion weekly active users, combined with its acknowledgment that demographic inequalities persist across income and geography, suggests user-composition tracking will remain part of its public reporting.
On the infrastructure side, OpenAI’s framing of 8 gigawatts “identified” toward a 30-gigawatt target by 2030 represents a more guarded posture than the rapid-fire partnership announcements of 2025, and the company has not confirmed whether financing arrangements for its full stated compute pipeline are in place.
