·  ·  · 

ChatGPT App’s Meteoric Rise Stalls: Downloads Dip and Daily Usage Tumbles as Rivals Surge

October 18, 2025
ChatGPT App’s Meteoric Rise Stalls: Downloads Dip and Daily Usage Tumbles as Rivals Surge
  • Download Growth Slows: According to app-analytics firm Apptopia, ChatGPT’s mobile app download rate is decelerating – October 2025 downloads are on pace to be ~8.1% lower than September’s [1]. (By contrast, earlier this year downloads were climbing rapidly.)
  • Usage Metrics Down: U.S. user engagement has softened: average time spent per daily active user (DAU) is down 22.5% since July and sessions per user are down 20.7% [2]. These trends suggest Americans spend less time and open the app less frequently than a few months ago.
  • Still Huge Numbers: Even with slowing growth, ChatGPT remains a chart-topping app. Appfigures reports it was the #1 most-downloaded app worldwide in September (≈45 million installs) [3], and Sensor Tower notes it became the fastest app ever to 1 billion total downloads in mid-2025 [4]. In absolute terms, the app still adds millions of installs daily [5].
  • Competitors Rising: New AI chatbot apps are gaining ground. Google’s Gemini surged – hitting ~12.6 million U.S. downloads in Sept (up 45% from August) thanks to its “Nano Banana” image model launch [6]. Startups like Perplexity AI have spiked regionally (e.g. becoming the #1 free app in India, surpassing ChatGPT [7] [8]). By contrast, Anthropic’s Claude saw modest uptake (just ~157,000 global downloads in its first week [9]) and has already slipped down app charts.
  • Analysis – End of Novelty: App-intelligence experts say this cooling was expected. Apptopia notes the ChatGPT app’s “experimentation phase” is over and usage is settling into routine patterns [10]. Industry analysts echo this: Omdia’s chief AI analyst Bradley Shimmin predicted ChatGPT’s meteoric growth would naturally become “radically less meteoric” as users move past the novelty stage [11].
  • Implications for OpenAI: Only ~5% of ChatGPT’s 800 million weekly users are paying subscribers [12], so new growth is crucial for revenue. OpenAI is already pivoting: CEO Sam Altman highlighted that “more than 800 million people use ChatGPT every week” [13] while the company rolls out new features (like in-chat “apps” and agents [14]) and pursues diverse revenue streams (shopping tools, video services, hardware and government contracts) to meet its aggressive funding and spending plans [15].

ChatGPT App Growth Slows After Explosive Surge

ChatGPT’s mobile app — which rocketed to the top of iOS and Android charts after its 2023 launch — now shows signs of peaking. According to Apptopia, global download growth “slowed after April” of 2025 and has flattened in recent months [16] [17]. In practical terms, October 2025 downloads were tracking ~8% below September’s level [18]. (Importantly, this measures growth rate, not total installs — the app still brings in millions of new users daily, but at a tapering pace.)

At the same time, daily engagement metrics are easing. Apptopia finds U.S. users are spending significantly less time in the app than before. Average time per active user has fallen 22.5% since July, and average sessions per user have dropped 20.7% [19]. In other words, Americans are using ChatGPT less frequently and for shorter periods each day. These trends coincide with reports that total user churn has stabilized – meaning the app is no longer attracting as many casual, one-time users, and mainly retains its core user base [20].

Despite this cooldown, ChatGPT’s mobile presence is still enormous. Appfigures data shows ChatGPT was the world’s most-downloaded app in September 2025, at roughly 45 million installs across iOS and Android [21]. Sensor Tower likewise notes ChatGPT recently became the fastest app ever to reach 1 billion global downloads [22]. (For context, this eclipse of past records underscores how unprecedented ChatGPT’s rise has been.) So far in 2025, ChatGPT’s download lead is far above other chatbots. According to Appfigures estimates, its year-to-date installs (~410 million by early Oct) dwarf those of Google’s Gemini (~131 million) or Perplexity (~43 million) [23]. In absolute terms, ChatGPT still far outpaces competitors, but the rate at which it was adding new users has mellowed.

Insights from App Analytics Firms

Independent analytics firms have been tracking these shifts. Apptopia’s analysis (as reported by TechCrunch) explicitly warned the app’s “overall pace of growth is slowing” [24]. Appfigures intelligence shows the recent months’ downloads are plateauing: ChatGPT hit an all-time high of ~50 million downloads in May, then 45 million in September [25] [26], indicating a slight ebb from month to month. Meanwhile, Sensor Tower’s broader AI-app reports highlight the context: in mid-2025 ChatGPT had already achieved half a billion monthly active users (fastest ever) and 1 billion installs [27], but the share it commands is coming under pressure. For example, Sensor Tower noted that by late 2024, Google’s Gemini had captured a majority of AI-chatbot downloads in India, where ChatGPT’s share shrank below 50% [28].

These intelligence reports also flag competitive dynamics. Appfigures noted in early October that Gemini nearly unseated ChatGPT as the top app in September, and that if Gemini maintains its pace, Meta’s Facebook/Instagram apps might permanently cede top positions [29]. (In September, Gemini scored ~40 million downloads worldwide thanks to Google bundling Nano Banana, up sharply from ~9 million the previous month [30].) At the same time, Appfigures reiterates that ChatGPT’s install figures remain enormous by any standard. Its analyst called ChatGPT the “AI king,” cautioning that even a slight dip in raw downloads may simply reflect a maturing growth curve [31].

Analysis: Experts See ‘Maturing’ After Novelty

What’s behind this slowdown? Data analysts point to natural maturation and changing user patterns. Apptopia suggests ChatGPT’s “experimentation phase” is over [32]. In early days users were discovering the app and its novel capabilities, but now many have integrated it into everyday tasks or dropped out. As Apptopia observed, people may be using ChatGPT only “when they need it or remember to use it,” rather than endlessly at first [33]. If only the time-per-user metric were falling, one might argue users were simply getting more efficient. But since both time and session counts are down, Apptopia’s view is that usage is simply normalizing.

Industry analysts agree this was predictable. In mid-2023, after initial web-traffic declines, analysts noted ChatGPT’s “radical” rise would naturally be followed by a slowdown as the novelty waned [34]. Bradley Shimmin of Omdia put it bluntly: having captured mass attention, ChatGPT would see a “radically less meteoric” fall-off in usage once people tried out other AI tools [35]. In other words, there’s a larger AI ecosystem now. As one analyst quipped, “there are more generative AI fish out there than ChatGPT for users to explore” [36].

OpenAI itself has not directly commented on the mobile app’s download curve (TechCrunch reports it did not respond to media queries [37]). Instead, OpenAI leadership highlights the platform’s massive overall usage. At its October Dev Day, CEO Sam Altman announced “more than 800 million people use ChatGPT every week” [38], a figure encompassing all access channels, and celebrated the company’s rapid expansion of developer tools and new features (such as allowing third-party “apps” inside ChatGPT) [39]. The message: OpenAI is focusing on broad adoption and new capabilities even as the mobile app enters a more mature phase.

Competitive Landscape: Gemini, Claude, Perplexity and More

Part of the download slowdown story is greater competition. Google’s AI assistant Gemini (formerly Bard) has been heavily promoted and pre-installed on Android devices. The recent release of Google’s Nano Banana image model gave Gemini a notable bump: Sensor Tower data show Gemini soared into the top of app charts in September [40], and Apptopia observed Gemini “shot up to the top charts” after Nano Banana’s launch [41]. With Google integrating Gemini into its core apps (e.g. Android, Search) [42], some users may be diverting queries away from ChatGPT.

Startups are also gaining traction. Perplexity AI (a search/chat hybrid) has exploded in certain markets. In mid-October, Perplexity’s CEO announced on social media that Perplexity had become the #1 free app in India on both Google Play and Apple’s App Store [43] [44], overtaking ChatGPT and Gemini in that region. This surge is partly thanks to a partnership bundling Perplexity Pro with a major telecom plan, and to strong local appeal [45] [46]. Globally, Perplexity has amassed tens of millions of installs; one analytics summary puts India installs over 38 million by mid-Oct [47]. While India may be a standout case, it reflects how specialized AI tools can attract users by focusing on niche strengths (Perplexity emphasizes factual, cited answers and web search integration).

Meanwhile, other entrants nibble at the edges. Anthropic’s Claude had a very different trajectory: when Claude 3 launched on iOS in May 2024, it garnered only ~157,000 downloads in its first week – a fraction of ChatGPT’s initial uptake [48] – and quickly fell out of top app charts [49]. TechCrunch noted Claude “suffers from having more competition in the AI space than ChatGPT did when it initially launched” [50], citing Google’s broader AI push and aggregator platforms like Quora’s Poe. Other AI-chat apps (from Character.ai to various generic “AI Chatbot” apps) have also reached millions of installs by gaming app-store SEO [51], highlighting that users are exploring many options.

In sum, ChatGPT’s relative market share is under pressure even as its absolute numbers remain high. As Appfigures put it, growth continues but “a touch lower than August’s haul” [52] suggests we’re at the top of the hill.

User Behavior Shifts: From Novelty to Utility

Analysts say the data suggest a classic user lifecycle effect. ChatGPT saw a frenzy of early curiosity and discovery use – from students tackling homework to professionals automating tasks. But as OpenAI’s own usage study (Sep 2025) shows, conversations have broadened to everyday tasks and professions [53]. Users are “increasing their activity over time,” according to OpenAI’s economic research team [54] [55], but that growth is more steady now.

The mobile app, in particular, may see less novelty repeat usage. Anecdotally, some heavy users may have moved from experimenting on the app to more routine usage on desktop or third-party integrations, or simply hit a saturation point in how often they need it. (A mid-2023 analysis of ChatGPT web traffic found that a plunge in site visits coincided with summer breaks, and speculated many users may have migrated to the mobile app, which web analytics didn’t capture [56].) Now the reverse could be true: users saturated the app, then curbed usage or shifted to the web or other tools.

Another factor could be free-vs.-paid fatigue. ChatGPT Plus subscription remains the main monetization path for consumers, but only a small fraction of free users convert at current pricing. Heavy users sometimes complain about message or usage limits on the free tier, but with many free users already in the funnel, further aggressive upselling might backfire. So far OpenAI seems relying on adding value (e.g. image generation, extended memory) rather than penalizing free access. Still, if downloads slow, OpenAI may need new ways to monetize beyond the Plus tier.

Implications for OpenAI’s Strategy and Roadmap

For OpenAI, the app slowdown is both a warning and a pivot opportunity. Financially, ChatGPT’s subscriptions are enormous – OpenAI is reportedly pulling in ~$13 billion per year, 70% of it from people paying $20/month [57] – but most of its 800M weekly users remain non-paying. The slower growth rate means adding large numbers of new subscribers will be harder. That helps explain why OpenAI is diversifying aggressively. TechCrunch notes the company’s five-year plan involves new revenue lines like shopping and video tools, hardware devices (rumored), government contracts, and even selling GPU compute via its “Stargate” data center project [58].

Product-wise, OpenAI is doubling down on making ChatGPT indispensable. The Dev Day keynote showcased “apps” inside ChatGPT and advanced AI agents [59], suggesting more reasons to use the platform daily. While the mobile app may not see its former surge, integrating ChatGPT technology into wider ecosystems (Google, Microsoft, or embedded devices) could sustain its reach. Indeed, Microsoft’s Copilot, Meta’s AI, and other AI tools are part of the broader battleground – meaning OpenAI’s monetization may shift increasingly toward enterprise and partnerships.

In summary, the ChatGPT mobile app is transitioning from a viral sensation to a mature product. It still commands huge usage, but growth has plateaued in the face of rising competition and normalizing user behavior. OpenAI will need to innovate (and possibly market) to reignite momentum. As experts note, the easy growth is over [60] [61] – from here on, ChatGPT will have to stand on product depth, ecosystem integration, and smart monetization to stay on top.

Sources: Recent app intelligence analyses and industry reporting [62] [63] [64] [65] [66] [67] [68]. These include TechCrunch, Appfigures, Sensor Tower, Reuters, and other tech-media accounts of ChatGPT and AI app trends.

Deepseek R1 vs ChatGPT O3 Mini – The Ultimate AI Battle in 2025! 🏆🤖

References

1. techcrunch.com, 2. techcrunch.com, 3. appfigures.com, 4. sensortower.com, 5. techcrunch.com, 6. www.techi.com, 7. www.indiatoday.in, 8. www.moneycontrol.com, 9. techcrunch.com, 10. techcrunch.com, 11. aibusiness.com, 12. techcrunch.com, 13. techcrunch.com, 14. techcrunch.com, 15. techcrunch.com, 16. techcrunch.com, 17. techcrunch.com, 18. techcrunch.com, 19. techcrunch.com, 20. techcrunch.com, 21. appfigures.com, 22. sensortower.com, 23. www.businessinsider.com, 24. techcrunch.com, 25. appfigures.com, 26. appfigures.com, 27. sensortower.com, 28. sensortower.com, 29. appfigures.com, 30. appfigures.com, 31. appfigures.com, 32. techcrunch.com, 33. techcrunch.com, 34. aibusiness.com, 35. aibusiness.com, 36. aibusiness.com, 37. techcrunch.com, 38. techcrunch.com, 39. techcrunch.com, 40. www.techi.com, 41. techcrunch.com, 42. techcrunch.com, 43. www.moneycontrol.com, 44. www.moneycontrol.com, 45. www.indiatoday.in, 46. www.indiatoday.in, 47. www.indiatoday.in, 48. techcrunch.com, 49. techcrunch.com, 50. techcrunch.com, 51. techcrunch.com, 52. appfigures.com, 53. openai.com, 54. openai.com, 55. openai.com, 56. aibusiness.com, 57. techcrunch.com, 58. techcrunch.com, 59. techcrunch.com, 60. techcrunch.com, 61. aibusiness.com, 62. techcrunch.com, 63. appfigures.com, 64. techcrunch.com, 65. www.moneycontrol.com, 66. aibusiness.com, 67. techcrunch.com, 68. techcrunch.com

Technology News

  • What is AWS? How a Monday outage hit sites and apps
    October 20, 2025, 4:24 PM EDT. An outage at Amazon Web Services (AWS) disrupted access to popular sites and apps on Monday, including Snapchat, Roblox, Fortnite, Robinhood, and the McDonald's app. AWS is a leading cloud platform offering more than 200 services, from computing power and storage to databases and networking, used by countless businesses worldwide. The incident highlights how many online services rely on a single cloud provider for core infrastructure, and how a disruption can cascade across the internet. Users should monitor status dashboards, understand failover and redundancy strategies, and consider multi-cloud or regional designs to improve resilience during cloud outages.
  • Cloud Outages to Worsen as AI Adoption Expands, Says Tech CEO After AWS Disruption
    October 20, 2025, 4:22 PM EDT. An AWS DNS outage disrupted major carriers, banks, streaming services and apps, impacting more than 1,000 companies. Future Tech Enterprise CEO Bob Venero warns that outages will rise as AI capabilities expand in the enterprise. He notes a growing trend of repatriation from public cloud to colocation and on-prem deployments to mitigate risk. AWS continues to invest heavily in AI-focused data centers, including plans to spend billions in Pennsylvania and Georgia in 2025, reinforcing its dominant position in the cloud market. The incident underscores customer risk tolerance and the need for diversified architectures. As enterprises adopt more AI, cloud complexity grows, and providers face potential disruption, making on-prem and colocation options more attractive for resilience.
  • Apple Watch SE 3 vs Series 11: Why the SE 3 is the smarter pick for most people
    October 20, 2025, 4:20 PM EDT. After upgrading from the Apple Watch SE 2, the author argues that the Apple Watch SE 3 is the better pick for most people compared to the Series 11. Both wearables share the S10 processor and offer an always-on display; the SE 3 matches most health and fitness features, including the Sleep Score, at a much lower price. The big caveat: blood-pressure monitoring isn't available yet in Australia due to TGA delays, which undermines the appeal of the higher-end model. Battery life still favors the Series 11, but the SE 3 can go longer with low-power mode (nearly two days). Overall, for many users, the SE 3 delivers nearly the same performance at a substantially lower cost, with only a few features reserved for the Series 11.
  • Apple Watch Series 11 eases wearable anxiety with Workout Buddy and empathetic coaching
    October 20, 2025, 4:18 PM EDT. After switching from WearOS to the Apple Watch Series 11, the author finds wearable anxiety diminishes thanks to a new feature called Workout Buddy, an AI-powered coach that offers encouragement during short workouts. The piece argues that activity rings and constant health data can provoke stress, but these tools can also soften the experience and make tracking feel supportive rather than punitive. It also references research from UNC and Yale showing that smartwatch data can contribute to anxiety and pathological monitoring. The takeaway: well-designed feedback and empathetic coaching can transform wearables from pressure points into trusted wellness partners.
  • Tesla kicks off Magnificent Seven earnings season with AI-driven bets and robotaxi focus
    October 20, 2025, 4:16 PM EDT. Tesla is set to report earnings on Wednesday, launching the Magnificent Seven's Q3 cycle. After a rough first half of 2025, the EV maker has staged a late rally, up about 10% since January while the broader group flirted with gains. Investors are weighing near-term results against longer-term opportunities in autonomy, robots, and a potential pay plan for Elon Musk alongside a major stock purchase. Wall Street remains divided on valuation, with price targets below Friday's close despite strong deliveries and an improving lineup. Visible Alpha projections sit at roughly $26.6 billion revenue and $1.5 billion in net income, while many analysts are still calibrating forecasts for a new generation of cars. Tesla's performance-and any AI/autonomy updates-will help set the tone for the Magnificent Seven as earnings season proceeds, including Netflix.

Don't Miss