Gen Z AI Statistics 2026
By Axis Intelligence Research
Co-author: Sarah Mitchell | Last updated: August 18, 2026 | License: CC BY 4.0
Gen Z uses artificial intelligence more than any other American age group and expects worse outcomes from it than any other American age group. Two-thirds of U.S. adults under 30 use AI chatbots. Fourteen percent think AI will be good for society. Axis Intelligence Research measures that contradiction at 31.5 points on the Gen Z AI Paradox Score — the widest of any age band, and the number this page exists to explain.
Quick Answer
Sixty-six percent of U.S. adults aged 18–29 use AI chatbots and 31% use them daily, the highest rates of any age group, according to Pew Research Center data collected in February 2026. Yet only 14% expect AI’s twenty-year impact on society to be positive. Axis Intelligence Research scores this divergence at 31.5 points on the Gen Z AI Paradox Score (GAPS) — 1.85 times the all-adult reading of 17.0, and the widest gap on the age curve. Among employed Gen Z, 48% say AI’s workplace risks outweigh its benefits.
Key Findings
- Axis Intelligence Research finds the Gen Z AI Paradox Score (GAPS) at 31.5 points for U.S. adults aged 18–29 as of February 2026 — the widest engagement-versus-confidence divergence of any age band, computed from Pew Research Center’s comparable four-question instrument.
- Axis Intelligence Research finds the AI paradox inverts at 65+, where GAPS reads −1.0 points — older Americans expect slightly more from AI than they engage with, producing a 32.5-point spread across the age curve and identifying the paradox as a young-adult phenomenon rather than a general one.
- Employment among U.S. workers aged 22–25 in the most AI-exposed occupations stands about 19% below where it would be had it tracked less-exposed peers, as of June 2026, according to the August 2026 revision of Stanford Digital Economy Lab’s “Canaries in the Coal Mine” — up from a 15% shortfall at the July 2025 data vintage.
- Gen Z’s generative AI adoption grew just four percentage points in twelve months, to 51% weekly-or-more use, according to Gallup’s February–March 2026 Voices of Gen Z survey of 1,572 respondents aged 14–29 — while agreement that AI speeds up work fell ten points to 56%.
- Axis Intelligence Research finds a 39.4-percentage-point spread in Gen Z AI adoption across the European Union, from 83.5% of 16-to-24-year-olds in Greece to 44.1% in Romania, based on Eurostat’s 2025 ICT household survey — a wider national spread than any other digital behaviour Eurostat tracks for this cohort.
How Many Gen Z Use AI in 2026?
Adoption is high and has stopped climbing. That combination is the story.
Gallup’s Voices of Gen Z survey, fielded February 24 to March 4, 2026 with 1,572 respondents aged 14–29 and a margin of sampling error of ±3.6 percentage points, puts weekly-or-more generative AI use at 51% — up four points from 47% a year earlier. For a technology in its fourth year of mass availability, four points is a plateau, not a curve.
The frequency distribution matters more than the headline. Twenty-two percent of Gen Z use generative AI daily, 29% weekly, 11% monthly, 20% once every few months, and 19% never. Nearly two in five sit in that bottom bracket — occasional or absent. The commonly repeated claim that Gen Z has universally adopted AI does not survive contact with the distribution.
Gen Z AI Usage Frequency, 2025 vs 2026
| Frequency | 2025 | 2026 | Source |
|---|---|---|---|
| Daily | 19% | 22% | Gallup, Voices of Gen Z |
| Weekly | 28% | 29% | Gallup, Voices of Gen Z |
| Monthly | 13% | 11% | Gallup, Voices of Gen Z |
| Once every few months | 19% | 20% | Gallup, Voices of Gen Z |
| Never | 21% | 19% | Gallup, Voices of Gen Z |
Source: Gallup, Voices of Gen Z, fielded February 24–March 4, 2026 (n=1,572, ages 14–29).
Students out-use workers. Fifty-six percent of Gen Z K-12 students report weekly-or-more use against 48% of Gen Z adults — a gap that inverts the usual technology-adoption pattern, where employment and income raise usage.
Which AI Chatbots Does Gen Z Actually Use?
Pew Research Center’s Americans and AI 2026 study, based on 5,119 U.S. adults surveyed February 17–23, 2026, gives the cleanest tool-level breakdown for the 18–29 band. ChatGPT reaches 61%. Gemini follows at 30%, then Copilot at 17%, Meta AI at 14%, Claude at 10%, Grok at 9% and Character.ai at 7%.
| Chatbot | Ages 18–29 | All U.S. adults | Source |
|---|---|---|---|
| ChatGPT | 61% | 44% | Pew Research Center |
| Gemini | 30% | 24% | Pew Research Center |
| Copilot | 17% | 17% | Pew Research Center |
| Meta AI | 14% | 14% | Pew Research Center |
| Claude | 10% | 6% | Pew Research Center |
| Grok | 9% | 8% | Pew Research Center |
| Character.ai | 7% | 3% | Pew Research Center |
Source: Pew Research Center, Americans and AI 2026, surveyed February 17–23, 2026 (n=5,119).
Two entries deserve attention. Copilot and Meta AI show no youth premium at all — identical shares among under-30s and the general adult population, which tells you these are distribution-driven products riding Microsoft 365 and Meta’s apps rather than products young people chose. Character.ai, at 7%, is the only tool where the under-30 share more than doubles the adult average.
Sarah Mitchell’s read: The tool table is where the “Gen Z is AI-native” framing starts to leak. ChatGPT’s 61% is a default, not a preference — it is the model people name when asked to name a model. What would tell us something about genuine fluency is multi-tool use, and none of these instruments measure it. Until a survey asks which model a respondent reaches for on a coding task versus a writing task, we are measuring brand recall with extra steps.
What Gen Z Uses AI For
Search leads at 54% among 18-to-29-year-olds. Work tasks follow at 41% of employed young adults, image and video creation at 34%, entertainment at 33%, diet and fitness information at 30%, and medical advice at 27%.
The figure that has drawn the most external attention is emotional support: 20% of adults under 30 say they have used a chatbot for it, against 13% of those aged 30–49 and 4% of those aged 50–64. Companionship, asked separately, sits at 7%. Seventy-two percent of the under-30 group read AI summaries at the top of search results — the highest of any age band, and the clearest signal that AI has already been absorbed into the information layer whether or not anyone opted in.
What Is the Gen Z AI Paradox Score (GAPS)?
Every major 2026 survey has now reported the same shape without naming it: the people using AI most are the people expecting least from it. GAPS — the Gen Z AI Paradox Score — is an Axis Intelligence Research metric that converts that shape into a single comparable number for any age band.
Formula and Inputs
GAPS is the distance between how much a cohort engages with AI and how much benefit it expects from AI, expressed in index points.
Engagement Score (E) = mean(% who ever use AI chatbots, % who use AI chatbots daily) Outcome Confidence Score (C) = mean(% expecting AI’s 20-year impact on society to be positive, % expecting it to be positive for them personally) GAPS = E − C
All four inputs come from a single instrument — Pew Research Center’s Americans and AI 2026 survey, fielded February 17–23, 2026 — so every age band is measured with identical question wording and a common field period. A competent outsider can recompute every reading below from Pew’s published tables.
GAPS by Age Band, February 2026
| Age band | Ever use chatbots | Use daily | Positive for society | Positive personally | E | C | GAPS |
|---|---|---|---|---|---|---|---|
| 18–29 | 66% | 31% | 14% | 20% | 48.5 | 17.0 | 31.5 |
| 30–49 | 61% | 34% | 17% | 26% | 47.5 | 21.5 | 26.0 |
| 50–64 | 42% | 19% | 18% | 24% | 30.5 | 21.0 | 9.5 |
| 65+ | 23% | 7% | 15% | 17% | 15.0 | 16.0 | −1.0 |
| All U.S. adults | 49% | 24% | 16% | 23% | 36.5 | 19.5 | 17.0 |
Inputs: Pew Research Center, Americans and AI 2026 (February 17–23, 2026). E, C and GAPS calculated by Axis Intelligence Research; arithmetic independently verified in Python across two passes.
Reading the Number
The 18–29 reading of 31.5 is 1.85 times the all-adult reading of 17.0. More usefully, the score declines monotonically with age and crosses zero: at 65+ it reads −1.0, meaning older Americans hold marginally more confidence in AI’s outcomes than their engagement with it would predict. The full spread from youngest to oldest is 32.5 points.
That inversion is the finding. If AI pessimism were simply a function of unfamiliarity — the story the industry has told since 2023 — the oldest and least-engaged cohort would post the widest gap. It posts the narrowest, and with the wrong sign. Pessimism about AI outcomes in the United States is concentrated precisely where hands-on exposure is deepest.
A second reading, computed from Gallup’s separate instrument, points the same direction with a larger magnitude: 51% weekly-or-more use against 15% of employed Gen Z saying workplace benefits outweigh risks — a 36-point gap. Axis Intelligence Research reports this alongside GAPS rather than merging it. Gallup surveys ages 14–29 including K-12 students, asks about workplace risk rather than twenty-year societal impact, and uses a different frequency threshold. Averaging the two would produce a number with no defensible referent.
Scope
GAPS measures divergence, not causation, and it is anchored to U.S. survey data. The Engagement Score weights breadth and intensity equally, an editorial choice rather than an empirically optimised one; a version weighting daily use more heavily would widen the 30–49 reading relative to 18–29, since that band posts the highest daily rate at 34%. The score is designed for year-over-year comparison on a stable instrument, and Pew changed the chatbot-use question wording in 2026, so the February 2026 reading is the baseline against which future readings will be measured.
Sarah Mitchell’s read: The temptation with a number like this is to treat it as a verdict on the technology. It is not. It is a verdict on the distance between what these products do and what they were promised to do, measured among the people best positioned to notice. Gallup’s own crosstab makes the point sharply — daily Gen Z users are meaningfully more positive than non-users, with 44% excited against 4%, yet even that group’s excitement fell 18 points in a year. The people who know the tools best are not getting more enthusiastic as the models improve. That is a product signal, not a sentiment signal.
Why Is Gen Z Turning Against AI?
Sentiment moved hard between the 2025 and 2026 Gallup waves, and it moved on the practical questions rather than the philosophical ones.
Excitement about AI fell 14 points to 22%. Hopefulness fell nine points to 18%. Anger rose nine points to 31%. Anxiety held steady at 42%. Curiosity, added as a response option in 2026, became the most common emotion at 49% — which the Walton Family Foundation, which commissioned the survey with GSV Ventures, framed as a generation still engaged but no longer sold.
| Emotion about AI | 2025 | 2026 | Change | Source |
|---|---|---|---|---|
| Excited | 36% | 22% | −14 pts | Gallup |
| Hopeful | 27% | 18% | −9 pts | Gallup |
| Angry | 22% | 31% | +9 pts | Gallup |
| Anxious | 41% | 42% | +1 pt | Gallup |
| Curious | — | 49% | new option | Gallup |
Source: Gallup, Voices of Gen Z, 2025 and 2026 waves.
The utility numbers fell alongside the emotional ones. Agreement that AI tools help complete work faster dropped ten points to 56%. Agreement that AI helps them learn faster dropped seven points to 46%. And 80% of Gen Z say it is very or somewhat likely that using AI tools will make it harder for them to learn in future.
That last figure is the one worth sitting with. It is not a claim about job displacement or existential risk. It is a cohort reporting that the tool it uses most makes it worse at the thing it is currently doing — and using it anyway.
Familiarity still correlates with warmth. Among daily Gen Z users, 69% report curiosity, 44% excitement and 38% hopefulness, against 28%, 4% and 2% among those who never use AI. Anxiety runs at 28% among daily users and 60% among non-users; anger at 18% and 59%. The exposure effect is real. It has simply stopped compounding.
How Is AI Affecting Gen Z Entry-Level Jobs?
This is where survey sentiment meets payroll records, and the payroll records are less ambiguous than the surveys.
Stanford Digital Economy Lab’s August 2026 revision of “Canaries in the Coal Mine”, built on ADP administrative payroll data, reports that employment among workers aged 22–25 in the most AI-exposed occupations stands roughly 19% below where it would be had it kept pace with similarly aged workers in less-exposed occupations, measured to June 2026. At the July 2025 data vintage the same measure read 15%. In levels, employment for that age group in the two most exposed quintiles fell about 11% between November 2022 and June 2026, while the three least-exposed quintiles grew about 10%.
Experienced workers show no comparable gap. The authors attribute the adjustment primarily to reduced hiring rather than increased separations, and are explicit that these are descriptive patterns rather than causal estimates of AI’s effect.
Independent administrative evidence points the same way. A U.S. Census Bureau working paper by Lee Tucker, You’re (not) Hired: Artificial Intelligence and Early Career Hiring in the Quarterly Workforce Indicators (CES-26-27, April 2026), finds regression-adjusted employment of workers aged 22–24 in the most AI-exposed quintile of industry-state cells declined 12% over the ten quarters following ChatGPT’s release, with an immediate 9% drop in hires as the primary driver.
The broader entry-level picture is consistent without being catastrophic. The Federal Reserve Bank of New York reports recent college graduate unemployment at about 5.6% in the second quarter of 2026, with underemployment at 42%.
| Indicator | Value | As of | Source |
|---|---|---|---|
| Employment shortfall, ages 22–25, most AI-exposed occupations | 19% | June 2026 | Stanford Digital Economy Lab |
| Same measure, prior vintage | 15% | July 2025 | Stanford Digital Economy Lab |
| Early-career employment change, ages 22–24, most-exposed quintile | −12% | Q2 2025 | census.gov (CES-26-27) |
| Early-career hires change, most-exposed cells | −9% | Q2 2025 | census.gov (CES-26-27) |
| Recent college graduate unemployment rate | 5.6% | Q2 2026 | Federal Reserve Bank of New York |
| Recent college graduate underemployment rate | 42% | Q2 2026 | Federal Reserve Bank of New York |
The counterweight belongs in the same section, because leaving it out would misrepresent the data. The U.S. Bureau of Labor Statistics projects data scientist employment growing 33.5% between 2024 and 2034, software developers growing 15.8% — an increase of more than 267,000 jobs — and information security analysts, actuaries, operations research analysts and computer and information research scientists each growing at least 19.0%. AI is contracting the number of entry doors into some occupations while expanding the occupations themselves.
Sarah Mitchell’s read: Two administrative datasets, built from different sources with different methods, both find an early-career hiring effect concentrated in AI-exposed work. That convergence is worth more than any single survey in this article. But the Stanford authors themselves flag that their gaps narrow once education is controlled for, and that some divergence predates generative AI — and I would rather cite them accurately than louder. The honest reading is that the entry rung is thinner, not that the ladder is gone. A 22-year-old reading the 19% figure should understand it describes hiring into exposed occupations, not the odds of finding work.
Do Schools Give Gen Z Guidance on AI?
Policy infrastructure moved faster in the last year than adoption did — a rare case of institutions catching up rather than lagging.
Among Gen Z K-12 students, 74% report that their school has AI rules, up from 51% in 2025. Sixty-five percent say they are permitted to use AI for schoolwork, up from 55%. Access to AI tools from school computers rose from 36% to 49%. Confidence followed: 56% now agree they will have the skills to use AI in daily life after graduation, up 12 points.
| School AI measure | 2025 | 2026 | Source |
|---|---|---|---|
| School has AI rules | 51% | 74% | Gallup |
| Permitted to use AI for schoolwork | 55% | 65% | Gallup |
| AI tool access on school computers | 36% | 49% | Gallup |
| Expect to be AI-ready after graduation | 44% | 56% | Gallup |
| School provides AI tools for schoolwork | — | 28% | Gallup |
Source: Gallup, Voices of Gen Z, 2025 and 2026 waves.
The 28% figure is the one that does not fit the trend. Three-quarters of students now attend schools with AI rules; roughly one-quarter attend schools that actually supply AI tools. Permission has scaled; provision has not. Meanwhile 52% of students expect to need AI skills for postsecondary study and 48% for their future careers.
How Does Gen Z AI Adoption Compare Across Europe?
U.S. surveys dominate this topic, which makes the European picture the most under-reported segment in the 2026 data — and the one where national variation is largest.
Eurostat’s 2025 ICT household survey puts generative AI use in the previous three months at 63.8% among 16-to-24-year-olds in the European Union, against 32.7% of the general 16–74 population — a multiple of 1.95 calculated by Axis Intelligence Research.
The national range is the finding. Greece leads at 83.5%, followed by Estonia at 82.8% and Czechia at 78.5%. At the other end, Romania records 44.1%, Italy 47.2% and Poland 49.3%. Axis Intelligence Research measures the spread at 39.4 percentage points — meaning a Greek 20-year-old is nearly twice as likely to have used generative AI as a Romanian one of the same age.
| Country | Gen AI use, ages 16–24 | Source |
|---|---|---|
| Greece | 83.5% | Eurostat |
| Estonia | 82.8% | Eurostat |
| Czechia | 78.5% | Eurostat |
| EU average | 63.8% | Eurostat |
| Poland | 49.3% | Eurostat |
| Italy | 47.2% | Eurostat |
| Romania | 44.1% | Eurostat |
Source: Eurostat, dataset isoc_ai_iaiu, 2025 survey wave.
Purpose splits sharply by age. Among EU 16-to-24-year-olds, 44.2% use generative AI for private purposes and 39.3% for formal education, against 9.4% of the general population for education. Work use is nearly identical across the two groups — 15.8% versus 15.1% — because most of this cohort has not entered the labour market yet. That last comparison is the quiet one: Europe’s Gen Z AI habit is currently an education habit, and it will convert to a workplace habit on a schedule nobody has modelled.
Methodology
Axis Intelligence Research assembled this dataset under the two-pass research-and-writing process described in our editorial standards. No figure on this page originates in model memory.
Research pass (August 18, 2026). Eight primary sources were fetched and read in full during production: Gallup’s Voices of Gen Z 2026 release, the Walton Family Foundation’s accompanying announcement, Pew Research Center’s Americans and AI 2026 age chapter, the Stanford Digital Economy Lab’s August 2026 “Canaries in the Coal Mine” revision, U.S. Census Bureau working paper CES-26-27, the U.S. Bureau of Labor Statistics Economics Daily release on AI and employment projections 2024–34, Eurostat’s February 2026 release on dataset isoc_ai_iaiu, and the Federal Reserve Bank of New York’s 2026:Q2 recent-graduate labour market update. Every quantitative claim in the prose above corresponds to a row in the accompanying CSV, with source organisation, document, URL, as-of date and retrieval date.
Proprietary metric. The Gen Z AI Paradox Score (GAPS) is an original Axis Intelligence Research construction. Components, weights and inputs are disclosed in full in the GAPS section above. Both the Engagement and Outcome Confidence scores are unweighted means of two Pew inputs each; GAPS is their difference. Arithmetic was verified in Python across two independent passes using different formulations of the same expression.
Compatibility rule applied. Gallup (ages 14–29) and Pew (ages 18–29) use different age spans, field periods and question wording. Their figures appear side by side throughout and are never averaged, blended or presented as a combined statistic. Eurostat’s “used in the last three months” measure is not comparable to Pew’s “ever used” measure and is reported only within its own section.
About This Dataset
Title: Gen Z AI Statistics 2026 — Axis Intelligence Research Compiled Dataset
Contents: 157 observations covering generative AI adoption frequency, chatbot tool share, use cases, sentiment, workplace trust, school AI policy infrastructure, early-career employment exposure and European national adoption, with the Gen Z AI Paradox Score and its components.
Geography: United States and European Union, with six EU member-state readings.
Temporal coverage: 2022–2034 (survey observations 2025–2026; employment projections to 2034).
Creator and publisher: Axis Intelligence Research.
Licence: CC BY 4.0 — free to reuse, redistribute and adapt with attribution. Format: UTF-8 CSV, one row per observation, fourteen provenance columns.
Canonical URL: https://axis-intelligence.com/gen-z-ai-statistics/
Citation Formats
APA: Axis Intelligence Research. (2026). Gen Z AI statistics 2026: Adoption, sentiment and the entry-level job gap [Data set]. CC BY 4.0. https://axis-intelligence.com/gen-z-ai-statistics/
MLA: Axis Intelligence Research. “Gen Z AI Statistics 2026: Adoption, Sentiment and the Entry-Level Job Gap.” Axis Intelligence Research, 18 Aug. 2026, axis-intelligence.com/gen-z-ai-statistics/.
Chicago: Axis Intelligence Research. “Gen Z AI Statistics 2026: Adoption, Sentiment and the Entry-Level Job Gap.” Axis Intelligence Research, August 18, 2026. https://axis-intelligence.com/gen-z-ai-statistics/.
Attribution block for reuse: “According to Axis Intelligence Research, the Gen Z AI Paradox Score (GAPS) reached 31.5 points for U.S. adults aged 18–29 in February 2026 — 1.85 times the all-adult reading of 17.0, and the widest divergence between AI engagement and expected benefit of any age band. Formula, inputs and downloadable data at axis-intelligence.com/gen-z-ai-statistics/.”
Frequently Asked Questions
Does Gen Z use AI more than any other generation?
On breadth, yes. Sixty-six percent of U.S. adults aged 18–29 use AI chatbots, ahead of 61% among 30-to-49-year-olds and 42% among 50-to-64-year-olds (Pew Research Center, February 2026). On daily intensity the picture flips: 34% of 30-to-49-year-olds use chatbots daily against 31% of under-30s. Gen Z is the broadest adopter; the cohort ahead of it is the heaviest.
Has Gen Z AI adoption stopped growing?
Effectively. Weekly-or-more generative AI use among Gen Z rose four percentage points in twelve months, from 47% to 51% (Gallup, February–March 2026). Nineteen percent still report never using it. Adoption has settled into a plateau rather than continuing the steep 2023–2024 curve.
Is AI actually taking entry-level jobs from Gen Z?
The administrative evidence shows a hiring effect concentrated in exposed occupations rather than economy-wide displacement. Stanford Digital Economy Lab measures employment for 22-to-25-year-olds in the most AI-exposed occupations at about 19% below its counterfactual as of June 2026, driven mainly by reduced hiring. A Census Bureau working paper finds a 12% early-career employment decline in the most exposed industry-state cells. Both research teams describe these as descriptive patterns, not causal estimates, and the Stanford authors note the gaps narrow once education is controlled for.
Do Gen Z workers trust AI-assisted work?
No, and the margin is wide. Sixty-nine percent of employed Gen Z place more trust in work completed without AI, 28% in AI-assisted work, and 3% in work produced solely by AI (Gallup, 2026). Forty-eight percent say AI’s workplace risks outweigh its benefits, against 15% who say the reverse — a deterioration from 37% and 20% respectively a year earlier.
Which AI chatbot is most popular with Gen Z?
ChatGPT, used by 61% of U.S. adults aged 18–29, followed by Gemini at 30% and Copilot at 17% (Pew Research Center, February 2026). Claude reaches 10% and Character.ai 7% — the latter being the only tool whose under-30 share more than doubles its all-adult share.
Are schools teaching Gen Z how to use AI?
Schools are regulating faster than they are equipping. Seventy-four percent of Gen Z K-12 students say their school has AI rules, up from 51% a year earlier, and 65% are permitted to use AI for schoolwork. But only 28% say their school actually provides AI tools for their work (Gallup, 2026).
Which country has the highest Gen Z AI adoption?
Within the European Union, Greece, where 83.5% of 16-to-24-year-olds used generative AI in the three months before Eurostat’s 2025 survey, ahead of Estonia at 82.8% and Czechia at 78.5%. Romania records the lowest at 44.1%. The EU-wide figure for this age group is 63.8%.
What is the GAPS score and how is it calculated?
The Gen Z AI Paradox Score (GAPS) is an Axis Intelligence Research metric measuring the distance between a cohort’s AI engagement and its expected benefit from AI. Engagement is the mean of chatbot ever-use and daily-use rates; Outcome Confidence is the mean of the shares expecting AI to be positive for society and positive for them personally. GAPS is the difference. All four inputs come from Pew Research Center’s Americans and AI 2026 survey, so every age band is measured identically. The February 2026 reading for ages 18–29 is 31.5 points.
If Gen Z uses AI most, why is it the most negative about AI?
Because exposure and approval have decoupled. GAPS declines monotonically with age and turns negative at 65+, which rules out unfamiliarity as the driver of pessimism — the least-exposed cohort posts the narrowest gap. Gallup’s crosstabs point the same way: daily Gen Z users are far more positive than non-users, yet their excitement still fell 18 points in twelve months. The people closest to these tools are not becoming more optimistic as the models improve.
Related Axis Intelligence Research
- Gen Z Statistics covers the generation’s demographics, education and workforce position.
- Gen Z Spending Habits Statistics covers commerce behaviour and the purchase-influence data that sits downstream of the platform figures above.
- Gen Z Social Media Statistics
- AI Statistics
- Gen Z Workplace Statistics