AI Spend Tripled to $40B. Adoption Didn’t.
Marketers are pricing AI-surface advertising on user growth. The user growth stopped. The money moved to a paying minority that was already there.
Consumer AI spending tripled to $40 billion in 2026. The number of people using AI grew three points, from 61 percent to 64 percent. Menlo Ventures published both numbers on the same page, on September 16, and the gap between them is the story.
The AI advertising land grab is built on a reach story: more users, more inventory, more of your media budget chasing a bigger audience inside Gemini, Perplexity, and Meta AI. Menlo Ventures’ 2026 State of Consumer AI report, released September 16, tells a different story. US AI adoption moved three points in a year, 61 percent to 64 percent. Global consumer AI spending tripled, from $12 billion to $40 billion, in the same period. The math works one way: the 14 percent of AI users who pay $100 or more a month, a group that barely grew in size, now account for roughly 60 percent of all US consumer AI spend. IAB is forecasting 9.5 percent US ad spend growth for 2026, citing agentic AI adoption as a driver. That forecast assumes an expanding audience. The audience is flat. The committed minority is spending harder. Audit your next AI-surface media plan for a frequency floor before you buy against total user counts.
The Take
The consensus says AI platforms are becoming mass channels, so marketers should treat Gemini, Perplexity, and Meta AI placements like early Google Search or early Instagram: get in now while reach is cheap and the audience is still forming. The data says the audience already formed, and it stopped expanding in 2025. What changed in 2026 is how hard the existing base pays and uses the product, not how many new people showed up. Perplexity built an ad product on the reach assumption and pulled it in February, citing trust concerns, before it ever tested a concentration strategy. Buying AI-surface media for reach is buying yesterday’s audience at tomorrow’s price. Buy for frequency and spend intensity, or wait.
Buying AI-surface media for reach is buying yesterday’s audience at tomorrow’s price.
Also Worth Knowing
Anthropic opened Claude Commerce Agents to Shopify, Visa, Mastercard, and Accenture on September 3. The release gives AI agents a standard way to browse, compare, and check out on merchant sites without a human clicking through. It is infrastructure for an agentic-checkout economy that consumer trust data says is not ready, built ahead of the trust gap closing. Track which of your commerce partners adopt the protocol this quarter before committing engineering time to your own integration.
The global AI martech market is projected to grow from $28 billion in 2025 to $74.3 billion by 2031, with platform consolidation named as a primary driver, according to a September 11 market report. Fewer vendors will hold more of the stack, and switching costs will rise with each renewal cycle. Run a consolidation exposure audit on your current martech stack this quarter, before a vendor’s acquisition forces the migration on their timeline.
Under the Radar
The European Commission’s final guidance on Article 50(4) of the AI Act took effect August 2, and most US marketing teams have not read it. It requires disclosure whenever AI generates or manipulates images, video, audio, or text that could pass as authentic, including AI product imagery, AI-voiced testimonials, and AI-manipulated scenes with real people, if that content reaches an EU audience. Reach is the operative word: a US-only company with an EU-visible social ad is in scope. One exception applies: meaningful human review and editorial control, with a named human accountable for the final output. That exception creates a job: reviewer of record. Expect agencies to start selling human-review sign-off as a paid compliance layer within two quarters. Name a reviewer of record on every AI-generated asset that could reach Europe, this week.
Frameworks & Vocabulary
The Reach Illusion: treating a platform’s total user count as your addressable audience, when the economic and engagement value concentrates in a small, already-committed cohort. Example: a media plan built on "Gemini has billions of AI users" is chasing the Reach Illusion; a media plan built on "14 percent of AI payers drive 60 percent of AI spend" is chasing the real number.
The Practitioner Move
Three questions for your next AI-surface ad rep, before you sign: What share of served impressions goes to users active five or more times a week? Can you target by session frequency, not monthly actives? What is revenue per targeted user against platform-wide ARPU? If the rep cannot answer question two, you are buying the Reach Illusion. Hold the budget until they can.
Think About This: The AI economy is concentrating. Concentrated is a different business to sell into.
Forward this to the media buyer building next quarter’s plan around AI-platform user totals.
About mktg.ai
mktg.ai is the Creative Intelligence System for modern marketing. The platform unifies every creative asset, channel, and KPI in one place, connecting creative performance to spend in real time so teams can act at the layer consumers actually experience. Features include Ask mktg.ai for natural-language queries against your marketing data, and Daily Alerts AI for automatic anomaly detection that surfaces performance issues without waiting for weekly reports. On a $5M media budget, mktg.ai customers typically recover over $100,000 by reallocating 15 to 20 percent of spend toward higher-ROI creative within months. Learn more at mktg.ai.
Sources
Comments