a16z's Latest Consumer AI Report Finds a Split Market: 29 of the Top 50 AI Businesses by Consumer Spending Never Appear on Traffic Rankings

The consumer AI market has quietly split into two economies: one measured in visits, the other in money. That is the central finding of the seventh edition of a16z's Top 100 Gen AI Consumer Apps report, published October 5, which for the first time adds a third measuring stick — US consumer card-spend data from YipitData — alongside the series' usual Similarweb web-traffic and Sensor Tower mobile-MAU rankings. The result exposes businesses that thrive entirely off the traffic charts: of the top 50 vendors ranked by consumer spending, 29 do not appear on either the web or mobile traffic lists at all.

The headline numbers describe a market carried by very few shoulders. As of August 2026, only 4.5% of US consumers in YipitData's e-receipt panel held an active paid personal subscription to at least one of ChatGPT, Gemini or Claude — up from 2.1% a year ago, but still a sliver. Within that narrow paying base, spending follows a steep power law: the top 1% of payers accounted for 19.5% of all observed consumer AI spend, more than the bottom 50% of spenders combined (16.6%). Those top spenders put an average of $903 per month on personal cards for AI — a figure up 80% in the last 18 months — while the median AI payer spends $25 a month and has barely expanded that spend over time. Only 13% of users who pay for one AI product pay for even one other.

Who are the power spenders? a16z's data says they look less like mainstream consumers and more like prosumers — individuals buying software to build, automate and create. The top 1% disproportionately purchase automation and product-building tools such as n8n, fal, Manus and Nous Research's Hermes Agent, and over-index on creative tools like Higgsfield, Figma and HeyGen. Many of these products are structurally invisible to traffic metrics: n8n charges by workflow executions, so a single automation running every five minutes consumes quota around the clock without anyone opening a website; fal bills per unit of model output, such as per second of generated video. For such products, visiting a site and using the service are entirely different events.

The three-list intersection is tiny. Only seven companies appear on the web traffic, mobile MAU and consumer spend lists simultaneously: ChatGPT, Claude, Suno, Perplexity, Photoroom, Canva and Notion — three model companies, two AI-native applications, and two incumbents that added AI. ChatGPT is the only product that ranks first on all three. Meanwhile the traffic rankings themselves have ossified: just 11 first-time entrants appeared across the combined web and mobile lists, the fewest in the report's seven editions. a16z also excluded NSFW-oriented products from this edition — under the old methodology they would have occupied more than 20% of the web traffic list.

The report's own framing of the tension comes from consumer partner Olivia Moore: consumer AI usage is broad but shallow for almost everyone. Roughly half of US consumers say they use AI, yet only about 25% reach daily frequency. ChatGPT's dominance is unambiguous — its August web visits ran about twice Gemini's and six times Claude's, and its US paid subscriber base is roughly three times either rival's — but the more competitive race is for second place. Claude has now passed Gemini in US paid subscribers, and its subscriber mix is strikingly top-heavy: about 7.5% of Claude's subscribers pay for plans of $100 or more per month, versus roughly 1% for both ChatGPT and Gemini. (That top-heavy mix is consistent with what SemiAnalysis's separate subscription-value study found this week — Anthropic's premium tiers are where the money concentrates.)

Subscriptions were the industry's first answer to model costs, but a16z argues they cap the addressable market at the 4.5% who will pay — which is why new models are surfacing. OpenAI's advertising business has reached an estimated $1 billion annualized run rate as of August 2026, with shopping and travel queries flagged as the next expansion surfaces. The report also highlights personal agents as the category most likely to break the seat-based mold: products such as Instinct, Tomo, Poke, Lindy and Town now report hundreds of thousands of users, incumbents have answered with Meta's Muse, OpenAI's Dots and xAI's Grok Bot, and iMessage has become the launch channel of choice — Instinct went viral there in August, and DoorDash has since shipped its own iMessage text-to-order agent. Agents, a16z notes, could monetize on transactions rather than seats, offering something closer to unlimited usage without a subscription.

The agent race is also churning fast. OpenClaw, the pioneer that defined the category and drove enthusiasts to buy Mac minis to run local instances, has fallen out of the rankings entirely as its traffic declined — its team was reportedly acquired by OpenAI, presumably to build assistant products. a16z partners Olivia Moore and Josh Elman, discussing the report on the firm's podcast, predicted the next edition will show mass adoption of consumer assistants, and argued the value is shifting to the software layer — with the categories that produced the internet's biggest companies, including social, dating, shopping and entertainment, still almost entirely open in AI-native form.

One caveat belongs on the record: the spend figures come from a US card-panel sample and are not company revenue — YipitData observes what American consumers put on personal cards, which misses enterprise, education and international spending entirely. But as a lens on who actually pays for consumer AI, the picture is unusually sharp: a market where the traffic leaderboard is nearly frozen, the money is concentrating in a prosumer elite, and the next growth lever — ads, agents, transaction fees — is aimed at everyone else.