Mobile is where everyday AI use is quietly exploding in the US. Between March and June 2025, Microsoft’s Copilot grew its US mobile reach much faster than rivals—even while OpenAI’s ChatGPT kept the biggest audience overall. This shift isn’t just about apps. It’s about distribution, defaults, and the pull of enterprise workflows people already live in.
Copilot’s 175% mobile surge and 5.6M new users
From March to June, Copilot’s US mobile audience jumped 175% to 8.8 million. In absolute terms, it added 5.6 million mobile users in that window—more than any other assistant measured.
ChatGPT still #1 by total users
ChatGPT remains the market leader by US mobile users, reaching roughly 25.4 million in June. Growth was steadier at about 17.9% over the period, which is normal for a product that got big early and now grows from a larger base.
Gemini’s momentum from Android/Pixel distribution
Google’s Gemini also gained speed on mobile—up about 68% to 14.3 million US users—helped by placement across Google surfaces and being preloaded on Pixel phones. Distribution still matters, especially when it sits near the search bar or camera.
Mobile up, desktop down
The platform shift is real. In the US, mobile AI usage rose to 73.4 million people between March and June (+5.3%), while desktop usage dropped 11.1% to 78.4 million. That gap is narrowing fast as assistants feel more “native” on phones.
Deep enterprise integration across Microsoft 365 and Edge
Why is Copilot accelerating on mobile? One big reason is how neatly it plugs into what companies already use—Microsoft 365, Edge, Teams, and Windows. When your calendar, files, and chats are already in that ecosystem, Copilot slides right in. Comscore’s leaders point to this productivity-first fit as a growth driver.
Distribution and default advantages
Copilot rides along with Microsoft’s footprint. Edge and Windows keep it close to the work. On mobile, Microsoft’s own apps and sign-in flows reduce friction, while rivals lean on search placement or device preloads. That little bit of “on by default” often decides which app gets opened first.
Familiar workflows and security assurances
For IT teams, “works with what we have” beats novelty. Identity, permissioning, and data controls carry over from Azure AD and Microsoft 365. That predictability lowers the risk of rolling AI into sensitive workflows, which nudges users to try the assistant on their phones, too.
Pricing and bundles that lower friction
Copilot shows up in enterprise plans, trials, and bundled experiences that feel like extensions rather than new software. This “it’s already there” effect reduces procurement hurdles and speeds up adoption across devices—especially for employees who don’t want another login.
85%+ of users stick with one assistant
Another important finding: more than 85% of US mobile users tend to stick with a single AI assistant. That means once people settle into a tool’s rhythm and shortcuts, they rarely hop. For product teams, this is a wake-up call—win the habit, win the user.
Implications for marketers and IT leaders
If your customers or employees build muscle memory in one assistant, cross-promoting a rival later is harder. Documentation, shortcuts, and prompts need to be designed around the assistant your audience actually uses. Training content and internal champions matter as much as feature lists.
Azure tops $75B on AI demand
The enterprise story shows up in the revenue, too. Microsoft says Azure surpassed $75 billion in fiscal-year revenue, up 34% year over year, with AI demand listed as a key driver across workloads. That scale helps fund the rapid Copilot rollout and the underlying model hosting.
OpenAI’s ARR hits $10B (and rising)
As of June, OpenAI reached about $10 billion in annual recurring revenue across ChatGPT and the API business. Reports in late July suggested annualized revenue around $12 billion, underscoring how quickly the category is expanding.
Agency dynamics and the Microsoft–OpenAI relationship
There’s also a channel story. Microsoft’s broad enterprise deals can give agencies and brands deep access to OpenAI models through Azure and Copilot, sometimes reducing the need for separate direct contracts. That’s reshaping who signs with whom—and why.
Use cases that are naturally mobile
Many assistant tasks are “small bites” that fit into moments between meetings: drafting replies, summarizing documents, capturing photos for on-the-go analysis, or voice notes turned into clean text. As these feel smoother on phones, mobile usage climbs.
Bundling with devices and plans (Motorola x Perplexity as a signal)
Hardware tie-ups are becoming a serious lever. Perplexity’s global partnership with Motorola puts an AI assistant directly on new phones with a trial of the paid tier. Expect more OEM and carrier bundles—whoever wins the home-screen slot often wins usage.
Why US numbers lead strategy decisions
The data discussed here is US-focused, and that matters. US enterprise software budgets, app store rankings, and OEM partnerships often set patterns that spread globally. If you operate internationally, watch the US signal for timing, then localize for data protection rules and device mixes.
Copilot for Microsoft-centric teams
If your company runs on Microsoft 365, Copilot tends to feel “built-in.” The learning curve is lower, and governance can ride existing policies. Employees can pull context from Outlook, OneDrive, Teams, and SharePoint without juggling multiple apps. That convenience is a big reason behind Copilot’s mobile climb.
ChatGPT for breadth and rapid experimentation
ChatGPT still offers the widest user base and a massive third-party ecosystem of prompts, plugins, and tutorials. For creators, analysts, and startups moving fast, that community can be a superpower—even if its mobile growth rate has normalized from early surges.
Gemini for Google-first and Android workflows
If your world runs on Google Workspace or you’re deep in Android features like Lens and voice capture, Gemini may be the smoother fit. Its position across Google surfaces and Pixel preload helps it show up at the exact moment of need.
Panel data vs. installed base
Remember these are usage estimates from a large, privacy-safe panel across web and apps—not device install counts. That’s great for trend lines but not the same as total downloads or MAUs reported by companies themselves.
Growth rates vs. absolute share
Copilot’s 175% growth is eye-catching, but it grew from a smaller base. ChatGPT, with a far larger audience, can post slower percentage gains while still adding huge numbers of people. Both things can be true at once.
Impact of preloads, defaults, and policy shifts
The fastest way to change behavior is to change defaults. Preloads, browser integrations, and single sign-on matter as much as model quality for everyday users. Watch OEM and carrier deals, app store placements, and workplace policy updates—those often explain sudden step-changes in usage.
30-day audit
Map your organization’s tools, data locations, and security requirements. Identify the top 10 workflows that could benefit from AI on mobile—email triage, meeting notes, approvals, field reporting, and sales follow-ups. Rank by impact and ease. Interview five power users in each department.
60-day pilot
Run a small pilot with one assistant per team to avoid confusion. Measure task time saved, error rates, and user satisfaction. Document prompt patterns that work. Add short “how we use it here” guides inside your internal wiki. Keep legal and security in the loop for data-handling guardrails.
90-day rollout
Scale to more users with training and office hours. Bake assistant links into your intranet, templates, and mobile device management profiles. Appoint champions in each department and keep a running “wins” list to share every two weeks. If you’re Microsoft-centric, Copilot may win by default; if you’re mixed, keep two assistants but standardize documentation.
Conclusion
The US mobile AI race is tightening. Copilot grew the fastest this spring, thanks to enterprise fit and smart distribution. ChatGPT still leads by audience size, and Gemini is riding Android momentum. The lesson is simple: defaults, bundles, and workflow fit matter as much as model horsepower. If you choose the assistant that sits closest to your people’s daily work, adoption follows—and on mobile, that adoption is happening faster than most teams expect.
FAQs
Q1. Is Copilot actually larger than ChatGPT in the US?
No. Copilot grew faster on mobile from March to June, but ChatGPT still has the largest US mobile audience overall.
Q2. Does the data include apps and mobile web?
Yes. The measurements track both mobile web and native app usage for AI tools across a large US panel.
Q3. Why is mobile AI use rising while desktop falls?
Assistants now fit quick, on-the-go tasks like voice dictation, photo input, and short replies. That makes phones the natural place to use them.
Q4. What’s the enterprise angle behind Copilot’s rise?
Copilot plugs into Microsoft 365, Edge, and Azure security models. For companies already on Microsoft, that lowers friction and speeds adoption.
Q5. How fast is the business side growing for the major players?
Microsoft reported Azure revenue above $75B for the year, driven in part by AI demand. OpenAI said it reached $10B ARR in June, with later reports placing annualized revenue around $12B by late July.
Read more blogs: Alitech Blog
Zeeshan Ali Shah is a professional blog writer at AliTech Solutions, and Realancer renowned for crafting engaging and informative content. He holds a degree from the University of Sindh, where he honed his expertise in technology. With a keen eye for detail and a passion for staying up-to-date on the latest tech trends, Zeeshan’s writing provides valuable insights to his readers. His expertise in the tech industry makes him a sought-after writer, and his work at AliTech Solutions has earned him a reputation as a trusted and knowledgeable voice in the field.
Leave a Reply