OneLittleWeb study: AI tools drew 144.5B visits, with traffic in Top 100

AI tool usage on the open web is now large enough to measure like a category, not a curiosity. OneLittleWeb analyzed 9,531 AI tools across 170+ categories and found 144.5 billion estimated web visits over the past 12 months, up 40.12% year over year.

The more important signal is concentration: the Top 100 AI tools captured 129.6 billion visits, or 89.69% of all AI tool web traffic tracked. In other words, most “AI adoption” that shows up in web behavior is increasingly routed through a relatively small set of default destinations.

The company outlined the dataset and methodology in its AI tools traffic study.

Table of contents

Jump to each section:

• What the traffic numbers say about AI’s “default destinations”
• ChatGPT’s scale advantage is a distribution advantage
• Category shift: chatbots are absorbing single-purpose tools
• Geography: intensity vs volume tells different adoption stories
• What this means for marketers

What the traffic numbers say about AI’s “default destinations”

OneLittleWeb’s comparison helps separate two truths that often get blurred together.

AI tools generated 144.5 billion web visits in the last 12 months measured, while OneLittleWeb’s broader dashboard of 10,173 major websites logged 3.44 trillion visits in the same period. That puts AI tools at 4.20% of tracked web traffic.

Strategic observation: AI can feel ubiquitous in daily work while still being a small slice of total internet attention. Both can be true, and marketers should plan for both.

Within the AI tools universe, concentration is the defining pattern. The Top 100 alone grew 49.94% year over year and captured 89.69% of visits tracked across 9,531 tools.

Strategic observation: When a category concentrates this fast, “brand awareness” matters less than “default selection.” The winners become habits.

There is also an important measurement caveat embedded in the methodology: these are estimated web visits, not unique users, revenue, subscriptions, app usage, or API usage. For marketing teams, that is not a weakness so much as a lens. Web visits are often where discovery, comparison, and “good enough” decisions show up first.

ChatGPT’s scale advantage is a distribution advantage

The largest single number in the dataset is ChatGPT’s traffic: 64.7 billion web visits, representing 44.76% of the entire AI tools market as measured here. Canva ranked second with 10.5 billion visits, which means ChatGPT drew more than six times the traffic of the runner-up in the overall ranking.

Growth is part of the story, too. ChatGPT added 26.6 billion web visits and grew 69.80% year over year. Gemini added 5.6 billion visits (450.33% growth), Claude added 2.4 billion (250.07% growth), and Grok grew 731.26% to reach 2.5 billion visits.

A common assumption is that “more competition” automatically reduces the leader’s leverage. The contrasting reality in this dataset is that challengers can grow quickly while the leader still expands its lead in absolute terms. The strategic implication is uncomfortable but practical: distribution and habit can matter more than feature parity.

For marketers, this is less about picking a “best” model and more about anticipating where audiences and internal teams will naturally congregate. If one interface becomes the default place people ask questions, draft copy, or explore options, it becomes a new kind of demand capture layer.

Strategic observation: In AI, the interface is becoming the channel.

Category shift: chatbots are absorbing single-purpose tools

The study highlights a category-level reallocation of attention.

Chatbots grew from 45.1 billion to 86.0 billion visits, up 90.82%, reaching a 59.52% market share. Translators moved the other way, losing 1.6 billion visits (down 22.52%).

At the tool level, Google Translate lost 1.5 billion visits (down 23.46%), the largest absolute decline among all tools in the ranking. Chegg declined 61.45%, Quizlet fell 29.09%, and Quillbot lost 25.24%.

It is tempting to interpret this as “users stopped translating” or “users stopped studying.” A more precise read is that users may be consolidating tasks into general-purpose assistants that can translate, explain, summarize, and generate in a single session.

Strategic observation: Single-purpose tools are being judged against a bundle, not a feature.

This is a positioning warning for marketing, too. If your product’s differentiation is “we do one thing very well,” you may still win, but you will likely have to justify why the workflow should leave the general assistant. That justification usually needs to be about trust, accuracy, governance, domain depth, or integration into real work systems.

Geography: intensity vs volume tells different adoption stories

OneLittleWeb separates two different notions of “AI leadership” by comparing visits per 100,000 people (intensity) versus total traffic (volume).

Singapore ranked first by intensity with 16.1 million AI web visits per 100,000 people, despite a population of 5.9 million. The Netherlands followed with 9.8 million visits per 100,000 people. The United States was at 8.4 million visits per 100,000 people, while still leading in total volume with 29.2 billion visits.

For marketers, this is a reminder that “where AI is hottest” depends on what you mean by hottest.

• If you are planning pilots, partnerships, or early adopter programs, intensity can point to places where AI usage is deeply embedded in daily behavior.
• If you are planning scale, volume still matters because it correlates with absolute reachable demand.

Strategic observation: Adoption intensity signals habit; adoption volume signals market size.

What this means for marketers

This dataset is not a campaign playbook, but it is a strong read on where digital attention is consolidating. The deeper shift is that AI is becoming a small number of high-frequency destinations, not a long tail of tools.

1. Treat “default tools” as part of your channel strategy
   When a handful of AI interfaces capture most visits, they shape discovery paths, brand comparisons, and even how users phrase their needs. Monitor how your category is described inside those environments.
2. Position against consolidation, not just competitors
   If chatbots are absorbing translators, writing tools, and study platforms, the competitive set is not “other tools like us.” It is the general assistant that is already open in the browser.
3. Use web-traffic signals to guide content priorities
   Web visits reflect top-of-funnel curiosity and repeat usage. If traffic concentrates around a few tools and categories, align your content, SEO, and creator partnerships around the workflows those tools normalize.
4. Don’t confuse fast growth with durable attention
   The study shows several tools growing quickly, but concentration suggests that many “AI moments” still route through a few leaders. Build strategies that work even if your favored tool’s share shifts.
5. Segment by geography using intensity and volume
   Singapore’s intensity and the U.S.’ volume imply different go-to-market plays. If you only look at totals, you may miss where AI behavior is most embedded.

The larger marketing implication is not “AI is big.” It is that AI attention is becoming more centralized even as the category grows. That tends to reward brands that understand default behaviors early: where users start, what they ask first, and what “good enough” looks like in a consolidated interface.

And because AI tools still represent 4.20% of tracked web traffic in this comparison, there is still runway. The next phase is likely less about whether AI gets adopted and more about which few destinations become the standard layer between intent and action.