India is the largest market for ChatGPT Images 2.0, OpenAI said Thursday, with roughly 5 million downloads of the ChatGPT app there during launch week compared to about 2 million in the United States, per Sensor Tower. The global picture is far more muted: app downloads rose 11% week-over-week, but daily active users and sessions ticked up only 1%. Similarweb pegged the bump in global web traffic at 1.6%.
The split between download growth and engagement growth is the story. New users are showing up to try the model, especially in emerging markets, but the existing base isn't using ChatGPT meaningfully more often after the upgrade. In India, daily active users rose 3.4% week-over-week — better than the global figure, but still modest given the download volume.
Pakistan, Vietnam, and Indonesia saw the sharpest spikes, with ChatGPT app downloads climbing as much as 79% week-over-week during the rollout, according to Sensor Tower. That points to Images 2.0 functioning as a new-user acquisition tool in markets where ChatGPT was less entrenched, rather than as a feature that pulls existing Western users back in.
Key facts
- 01ChatGPT logged about 5 million downloads in India during the launch week of Images 2.0, versus roughly 2 million in the US, per Sensor Tower.
- 02Global ChatGPT app downloads rose 11% week-over-week, but daily active users and sessions climbed only 1%.
- 03Pakistan, Vietnam, and Indonesia saw download spikes of up to 79% week-over-week during the rollout.
- 04Daily active users in India rose 3.4% week-over-week, per Similarweb; global web traffic rose 1.6%.
- 05Images 2.0 adds rendering for non-Latin scripts including Hindi and Bengali, plus a 'thinking' mode that generates multiple variations per prompt.
OpenAI built the release with non-English markets in mind. Images 2.0 improves rendering of non-Latin text, including Hindi and Bengali, and adds a 'thinking' capability that refines outputs and generates multiple variations from a single prompt. The Hindi and Bengali support in particular maps directly onto India's user base.
“ChatGPT was downloaded roughly 5 million times in India during launch week and about 2 million in the United States, yet global daily active users rose only 1% week-over-week.”— Jaeden Schafer
OpenAI said in its post that Indian users are leaning into self-expression rather than utility. "In India, the early trends suggest ChatGPT Images 2.0 is largely being used as a form of self-expression. Rather than purely functional outputs, users are creating studio-style portraits from everyday photos, social media-ready images, and imaginative visuals that place themselves at the center," the company wrote. Fantasy newspaper covers, tarot-style visuals, fashion moodboards, and restored old photos all came up as common uses.
The competitive backdrop matters. Google's original Nano Banana image model also saw outsized early traction in India, which suggests India is becoming the proving ground for consumer AI image tools more broadly. Whichever model wins mindshare among hundreds of millions of mobile users there sets pricing and feature expectations everywhere else.
For OpenAI, the geographic skew is a double-edged result. India delivers scale — the kind of raw user numbers that justify the engineering investment in Hindi and Bengali support — but monetization per user in India is structurally lower than in North America or Western Europe. A launch that lands hardest in Pakistan, Vietnam, and Indonesia is a launch that needs a longer runway to translate into revenue.
The thin global engagement lift is the more uncomfortable signal. ChatGPT's existing power users in the US and Europe didn't measurably ramp their usage when Images 2.0 dropped, which suggests image generation is no longer the differentiator it was a year ago. Image quality has converged enough across OpenAI, Google, and others that a new release moves downloads but not session counts.
That convergence is a problem OpenAI will hit again. The company is releasing into a market where each model upgrade has to fight harder for attention, even as competitors like Google's Nano Banana line and a wave of open-weight image models close the quality gap. The 11% download lift versus the 1% engagement lift is what feature-fatigue looks like in a chart.
Still, the India numbers are real, and they're not just about novelty. Sustained 3.4% daily active user growth in a market this large compounds. If OpenAI can convert even a fraction of those new Indian users into paying subscribers — or sell them through a lower-priced regional tier — the geographic mix of its revenue base starts to look very different by the end of 2026 than it did entering the year.
The takeaway from Images 2.0 is that OpenAI's growth story is increasingly written in markets where ChatGPT is still being discovered, not where it's already a fixture. That's healthy for the top of the funnel and awkward for the unit economics. The next test is whether OpenAI builds pricing and product features that match where its users actually are, rather than where its San Francisco engineering team assumes they'll be.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
