Paid vs Free AI: Who Uses It, Who Pays, Who Waits
By Jonathan Kvarfordt · Published · Updated
Bottom line
About half of US adults use AI chatbots, and roughly one in four adults use them daily (Pew, Feb 2026). That daily group is about half of chatbot users. Few households pay for AI directly: in PNC's transaction-data panel, 2.2% of households paid for a generative-AI subscription in May 2026; a16z highlighted the finding in September 2026. The people who use AI every day are going deeper fast. Many people who do not use it chose not to, and they are falling further behind the daily users. If you feel behind, the data says most people sit below you on the ladder. The people above you are pulling away.
No survey we found measures "feeling behind" directly. This page measures use, payment and attitude, and says where each number stops.
The numbers: paid vs free
Paying is rare at the household level.
- In PNC's transaction-data panel, 2.2% of households paid for a generative-AI subscription in May 2026. Subscribing households spent about $31 a month, up from about $22 two years earlier (PNC Consumer Health Check, June 15, 2026). PNC says its data "may have a degree of selection bias." It covers PNC customers' card spending only.
- By income, more than 4% of higher-income PNC households paid, about 2% of middle-income and less than 1% of lower-income (PNC, June 2026).
- PNC households that subscribe keep paying longer: 7 consecutive months on average, up from around 5 at the start of 2024 (PNC, March 2026).
- a16z wrote on Sep 30, 2026 that "As of April, barely ~2% of US households were paying for some AI service" and that "The number is higher now, and growing" (a16z, State of Markets II).

- In YipitData's US e-receipt panel, 4.5% of consumers had a paid personal ChatGPT, Gemini or Claude plan in August 2026, up from 2.1% a year earlier, according to a16z's Top 100 Gen AI Consumer Apps report (Oct 5, 2026). That is a purchase panel covering three products. It is not a national count.
- Bank of America Institute reported in March 2026 that "only around 3% of Bank of America households currently pay for AI services," based on its payments data. That is a separate bank panel with its own method.
Spending is concentrated. In the same YipitData panel, the top 1% of payers averaged $903 a month and accounted for 19.5% of observed consumer AI spend. The bottom half of payers accounted for 16.6%. The median payer spent $25 a month. Only 13% of people who pay for one AI product pay for a second (a16z, Oct 5, 2026; consumer cards only, work spending excluded).
Using AI for free is common.
- 49% of US adults say they use AI chatbots; 51% do not (Pew, survey of 5,119 adults, Feb 17-23, 2026).
- 61.8% of adults aged 18 to 64 use generative AI for any purpose, up from 44.6% in Q3 2024 (St. Louis Fed FRED, Real-Time Population Survey, Q2 2026).
- 64% of US adults have used AI, up from 61% a year earlier (Menlo Ventures with Morning Consult, 5,067 adults, July 2026). Menlo is a venture firm with portfolio companies in consumer AI.
Paying, as people report it. Menlo says 55% of AI users pay for at least one AI product. Among payers, 48% pay themselves, 34% have a family member or friend paying, and roughly 20% have an employer or school paying (select all). That is a survey of AI users, counting any AI product and any payer. It measures something different from PNC's card data, so the two numbers do not belong side by side.
Company-stated scale. OpenAI said on Feb 27, 2026 that ChatGPT has more than 900 million weekly active users and more than 50 million consumer subscribers, plus more than 9 million paying business users. These are global, company-stated figures. Anthropic has not published a subscriber count; a spokesperson told TechCrunch in March 2026 that Claude paid subscriptions "more than doubled this year."
Feel vs use
Worry holds steady as use climbs.
- 52% of US adults say the increased use of AI in daily life makes them more concerned than excited; 37% feel both equally and 9% feel more excited (Pew, June 22-28, 2026). In 2021, 37% said more concerned.
- 40% expect AI's effect on society over the next 20 years to be negative, and 16% expect it to be positive (Pew, Feb 2026; sums are our arithmetic).
- Over the same stretch, chatbot use rose from about a third of adults in 2024 to about half in 2026 (Pew).
People report concern and keep using it. Among non-users, the reasons are deliberate: 70% distrust AI-generated information (up from 58%), 76% cite privacy and 83% would rather deal with a person (Menlo, July 2026).
The adoption ladder
Each rung uses only sourced shares. Pew's frequency split covers all US adults in Feb 2026. No source splits the regular-use rungs into free and paid.

| Rung | Share | Source and limit |
|---|---|---|
| No use | 51% do not use AI chatbots | Pew, Feb 2026. Menlo puts adults with no AI use in six months at 36%. Different questions. |
| Occasional | 15% use chatbots less than several times a week | Pew, Feb 2026. Free or paid: no data. |
| Regular | 10% several times a week; 24% about once a day or more | Pew, Feb 2026. 24% of all adults is our sum of 4% almost constantly, 12% several times a day and 8% once a day. Pew's text calls it "roughly one-in-four" adults. Among chatbot users alone, that is about half (our arithmetic: 24 of 49). Free or paid: no data. |
| Paid personal | 2.2% of PNC households (May 2026) | PNC panel only. No nationally representative count of paying households exists in what we found. |
| Work power users | 15% of US employees use AI at work daily (Gallup Panel, n=22,573, May 6-20, 2026); 16% in Aug 2026 | Who pays for those seats: no data. |
The gap is widening
- Daily use among US adults rose from 19% to 25% in a year while total adoption rose three points (Menlo).
- AI payers use AI daily at nearly twice the rate of non-payers, 50% vs 26%. The 14% of payers who spend $100 or more a month account for 60% of consumer AI spending (Menlo).
- Worldwide, daily generative-AI use at work rose from 14% to 22% in a year (PwC Global Workforce Hopes and Fears 2026, 49,364 workers in 48 countries, published Sep 29, 2026). PwC also places 56% of workers in a group it says is "falling behind." That is PwC's own grouping, based on skills and access to training. Workers did not report feeling behind.
- At work, frequent use (a few times a week or more) rose from 11% in May 2023 to 32% in August 2026 (Gallup).
What it means if you feel behind
What X is saying. On X, a16z's Oct 1 post saying "98% of US households aren't paying for AI yet" drew nearly 6 million views. The most common reply to "I'm behind" posts points to the low payment figures: almost nobody pays. The main pushback says AI bundled into Google, Apple and Amazon plans goes uncounted, and free tiers are good enough. This is a read of posts, not a survey.
If you use AI a few times a week, you are ahead of most adults on Pew's ladder. If you use it daily, about a quarter of adults are with you. The people who chose to wait are falling further behind the daily users, and many of them waited on purpose: distrust, privacy and a preference for people.
So calibrate against the right group. Compare yourself to your industry and your competitors' teams. At work, daily use is 42% in technology, 27% in finance and 22% in professional services; the other seven industries Gallup tracks range from 9% to 15% (Gallup, May 2026). A daily habit puts a revenue leader in insurance, banking, manufacturing or hospitality ahead of most of the field.
The opportunity now for revenue leaders
Firms. 23.8% of US businesses used AI in a business function in the last two weeks (Census BTOS, reference period Sep 7-20, 2026). By sector: finance and insurance 40.1%, information 50.2%, professional and technical services 42.7%, health care 24.8%, manufacturing 20.8%, retail 16.8%, accommodation and food services 10.1%. Census changed this question in November 2025, so earlier periods do not compare.
Teams. Access runs ahead of habit. 47% of US employees say their organization has integrated AI; 30% use it a few times a week or more; 25% say their organization has communicated a clear AI plan (Gallup, May 2026).
Sellers. Gartner found AI saves sellers 4.8 hours a week, and 72% of sales organizations report low reinvestment of that time (survey of 210 sales leaders, Jan-Feb 2026). Organizations that reinvest are 2.2x more likely to exceed customer growth goals.
What to do:
- CRO: Set one weekly AI habit per role and a metric for the time it frees. Gallup and Gartner both point at habit and reinvestment as the gap.
- Marketing: Your buyers split. About half use chatbots, and many non-users distrust AI and prefer a person. Keep a human path in every journey, and assume part of your audience researches you through a chatbot.
- Sales: Decide where the 4.8 saved hours go: more customer conversations, better account plans. Track it.
- Operators: Pull your own team's use rate before you buy more seats. Count daily users, weekly users and non-users, by role. Re-run it each quarter.
Why we might be wrong
- Free tiers are strong. Paying may stay rare because free AI does the job for most people. Low household payment can coexist with heavy use.
- Bundled AI goes uncounted. AI that comes with Google, Apple or Amazon plans, phones and software does not show up as an AI subscription in card data. Payment shares from PNC, Bank of America and YipitData all miss it.
- Work spending is hidden too. PNC's count misses employer- or school-paid seats, shared family plans, AI bundled into phones and software, app-store billing that may not map to an AI merchant, non-PNC cards and API spend. Menlo reports roughly 20% of payers have an employer or school paying.
- Measurement differs by source. "Use AI chatbots" (Pew), "use generative AI" (St. Louis Fed RPS) and "used AI" (Menlo) are different questions with different results. Pew changed its question wording in 2026. Census BTOS changed its AI question in November 2025.
- Opting out can be rational. For many jobs and people, waiting costs little. The holdouts' distrust may prove well founded in some uses.
- Feeling behind is unmeasured. We found no survey that asks it. The thesis rests on use data. Use data shows the gap. It cannot show how people feel about the gap.
FAQ
What share of US households pay for AI? No national count exists in what we found. In PNC's transaction-data panel, 2.2% of households paid for a generative-AI subscription in May 2026; a16z highlighted the finding in September 2026. Bank of America put its own households at about 3% in February 2026, and YipitData's panel showed 4.5% paying for ChatGPT, Gemini or Claude in August 2026. Each is a separate panel.
How many US adults use AI for free? About half of US adults use AI chatbots (49%, Pew, Feb 2026). Pew does not split free and paid use.
How often do people use AI? About a quarter of US adults use AI chatbots once a day or more, which is about half of chatbot users. 10% of adults use them several times a week and 15% less often (Pew, Feb 2026).
How do Americans feel about AI? 52% say AI in daily life makes them more concerned than excited, and 9% more excited (Pew, June 2026).
Am I behind on AI? If you use AI a few times a week, you are ahead of most US adults by Pew's numbers. Compare yourself to your industry: daily AI use at work ranges from 9% to 42% by industry (Gallup, May 2026).
Get the weekly read on where AI is moving revenue. Subscribe to the Report.
Sources (all read Oct 12, 2026)
- PNC Economics Research, Consumer Health Check, June 15, 2026 (May 2026 data): https://www.pnc.com/content/dam/pnc-com/pdf/aboutpnc/EconomicReports/consumer_health_check/PNC_Research_Consumer_Health_Check_June_2026.pdf
- PNC Economics Research, Consumer Health Check, March 12, 2026: https://www.pnc.com/content/dam/pnc-com/pdf/aboutpnc/EconomicReports/EconomicUpdates/2026/PNC_Research_Consumer_Health_Check_March_2026.pdf
- a16z (David George), State of Markets II, Sep 30, 2026: https://a16z.com/state-of-markets-ii/
- Pew Research Center, American Trends Panel Wave 187 topline, Feb 17-23, 2026, n=5,119: https://www.pewresearch.org/wp-content/uploads/sites/20/2026/06/PI_2026.06.17_Americans-and-AI_TOPLINE.pdf
- Pew Research Center, "Americans have grown more concerned about AI over time," Aug 18, 2026 (survey June 22-28, 2026): https://www.pewresearch.org/chart/americans-have-grown-more-concerned-about-ai-over-time/
- St. Louis Fed FRED, Real-Time Population Survey GenAI adoption (Bick, Blandin, Deming), updated Aug 4, 2026: https://fred.stlouisfed.org/series/RPSGENAIUSAGESHAREALL and https://fred.stlouisfed.org/series/RPSGENAIUSAGESHAREWORK
- Gallup, Global Indicator: Artificial Intelligence, updated 2026: https://www.gallup.com/699797/indicator-artificial-intelligence.aspx
- Menlo Ventures, 2026: The State of Consumer AI, Sep 16, 2026 (Morning Consult, n=5,067, July 2026). Investor research: https://menlovc.com/perspective/2026-the-state-of-consumer-ai/
- US Census Bureau, Business Trends and Outlook Survey, National and Sector files, period 2026-20 (published Oct 8, 2026): https://www.census.gov/hfp/btos/data_downloads
- OpenAI, "Scaling AI for everyone," Feb 27, 2026. Company-stated: https://openai.com/index/scaling-ai-for-everyone/
- TechCrunch (Julie Bort), Mar 28, 2026, Anthropic spokesperson statement: https://techcrunch.com/2026/03/28/anthropics-claude-popularity-with-paying-consumers-is-skyrocketing/
- Gartner press release, May 19, 2026 (survey of 210 sales leaders): https://www.gartner.com/en/newsroom/press-releases/2026-05-19-gartner-survey-finds-ai-saves-sellers-nearly-five-hours-per-week-yet-seventy-two-percent-of-sales-organizations-fail-to-reinvest-time-in-high-value-activities
- a16z (Olivia Moore), Top 100 Gen AI Consumer Apps, 7th edition, Oct 5, 2026 (YipitData panel): https://a16z.com/100-gen-ai-apps-7/
- Bank of America Institute, "Not quite mAInstream," March 2026: https://institute.bankofamerica.com/economic-insights/consumer-ai-usage.html
- PwC Global Workforce Hopes and Fears 2026, press release Sep 29, 2026: https://www.prnewswire.com/news-releases/companies-risk-losing-their-most-ai-savvy-employees-while-leaving-majority-behind-pwc-302891362.html
- Gallup, Q2 2026 workplace AI update (n=22,573): https://www.gallup.com/workplace/712736/organizational-adoption-jumps-six-points.aspx
- Pew Research Center report, Jun 17, 2026: https://www.pewresearch.org/internet/2026/06/17/americans-and-ai-2026-chatbots-smart-devices-and-views-on-impact/
Could not confirm
- A national, representative count of US households paying for AI.
- The a16z chart value: the article text says "barely ~2%" as of April 2026.
- Any direct survey measure of feeling behind on AI.
How to cite this research
Written by Jonathan Kvarfordt, Founder and Principal Analyst, The Revenue AI Report. Published under CC BY 4.0.
APA
Kvarfordt, J. (2026). Paid vs Free AI: Who Uses It, Who Pays, Who Waits. The Revenue AI Report. Retrieved from https://www.therevenueaireport.com/research/paid-vs-free-ai-adoption-2026
MLA
Kvarfordt, Jonathan. "Paid vs Free AI: Who Uses It, Who Pays, Who Waits." The Revenue AI Report, 12 Oct. 2026, www.therevenueaireport.com/research/paid-vs-free-ai-adoption-2026.
BibTeX
@misc{kvarfordt2026paidvsfreeaiadoption2026,
author = {Kvarfordt, Jonathan},
title = {Paid vs Free AI: Who Uses It, Who Pays, Who Waits},
year = {2026},
publisher = {The Revenue AI Report},
url = {https://www.therevenueaireport.com/research/paid-vs-free-ai-adoption-2026}
}