Advertiser Disclosure: This guide may contain affiliate links. If you purchase through them, we may earn a commission at no extra cost to you. Every statistic comes from independent third-party research linked to its original source, and no tool was hands-on tested for this article.
- 97% of content marketing programs use AI, but only 1% of marketers say their work is 100% AI-generated (Siege Media, 2026, 353 marketers).
- 49% of U.S. adults use AI chatbots, up from 33% in 2024 (Pew Research Center, 2026, 5,119 adults).
- Generative AI users report saving 5.4% of their work hours, about 2.2 hours in a 40-hour week (Federal Reserve Bank of St. Louis, November 2024 survey).
- Estimates of AI-written web content run from 10% (Pew Research Center, July 2026) to 74.2% (Ahrefs, April 2025) because each study counted something different.
- Across 600,000 pages, Ahrefs found a 0.011 correlation between a page’s share of AI content and its Google ranking (2025).
Look for a number to back up a point about AI writing and you’ll see the same figures on a dozen pages. Many of those pages cite each other instead of the original study, and a few figures are hard to trace to any named study at all. That makes it hard to know which AI writing statistics are safe to put in a report or a client deck.
This page works differently. Each of the 45 statistics below links to the organization that produced it, with the year and the sample size where the source publishes one. Where credible sources disagree, you’ll see both figures and what each one measured.
The headline finding is a gap between use and automation. In Siege Media’s survey of 353 marketing professionals, 97% said their content program uses AI, yet only 1% said their work is 100% AI-generated. Among the wider public, 49% of U.S. adults now use AI chatbots, up from 33% in 2024, according to a Pew Research Center survey of 5,119 adults.
Key AI Writing Statistics at a Glance
If you only have a minute, these six figures cover adoption, time savings, and how much of the web AI now touches.
- 97% of content marketing programs use AI, up from 90% in 2025, yet only 1% of marketers say their work is 100% AI-generated. (Siege Media, 2026)
- 49% of U.S. adults use AI chatbots, and 24% use one every day. (Pew Research Center, 2026)
- 54.6% of U.S. adults ages 18 to 64 used generative AI in August 2025, and 37.4% used it for work. (Federal Reserve Bank of St. Louis, 2025)
- 5.4% of work hours is the average time generative AI users reported saving, about 2.2 hours in a 40-hour week. (Federal Reserve Bank of St. Louis, November 2024 survey)
- 74.2% of 900,000 newly published web pages contained AI-generated content, though only 2.5% were classified as fully AI-written. (Ahrefs, April 2025)
- 38% of content marketers use AI for editing, double the 19% of a year earlier, while use for drafting fell from 57% to 44%. (Siege Media, 2026)
Read together, the numbers describe assistance more than replacement. Most people who write for a living now use AI at some stage, and most of them still edit and finish the work themselves.
Choosing a tool? Compare your options in our complete guide to AI productivity tools.
What Is AI Writing? Adoption vs. Saturation
- Adoption counts people: the share of a group, such as marketers or students, that uses AI at all. Siege Media’s 97% is an adoption figure.
- Intensity counts how much of the work AI handles. In the same survey, the most common answer (48% of respondents) was that AI touches 11% to 40% of their work.
- Saturation counts published content: the share of pages or articles that show signs of AI text. Ahrefs’ 74.2% is a saturation figure, and 71.7% of the pages it analyzed mixed AI and human writing.
A team can reach 100% adoption and still publish work that is mostly human-written. Each statistic below names the group or the content sample it describes, so you can tell which of the three you’re looking at.
AI Writing Statistics by Category
The 22 statistics in this section cover who uses AI, how marketers apply it, how large the market is, and what it’s doing to search.
How Many People Use AI Writing Tools?
Start with the widest question, which is how many people use generative AI at all. Three large trackers give three different answers because each one defines its population differently.
- 49% of U.S. adults use AI chatbots such as ChatGPT, Gemini, or Copilot, up from 33% in 2024. (Pew Research Center, 2026, 5,119 adults)
- 24% of U.S. adults use a chatbot every day, including 4% who say they use one almost constantly. (Pew Research Center, 2026)
- 54.6% of U.S. adults ages 18 to 64 used generative AI in August 2025, up from 44.6% a year earlier. (Federal Reserve Bank of St. Louis, 2025)
- 37.4% used generative AI for work in the same Fed survey, up from 33.3%. Pew’s separate survey puts work use at 38% of employed adults.
- 5.7% of all U.S. work hours involved generative AI in August 2025, up from 4.1% in November 2024. (Federal Reserve Bank of St. Louis, 2025)
- 57% of U.S. adults under 50 use ChatGPT, compared with 28% of those 50 and older. (Pew Research Center, 2026)
- 17.8% of the world’s working-age population used a generative AI product in the first quarter of 2026, up from 16.3% in late 2025. (Microsoft AI Economy Institute, 2026)
- 31.3% of working-age people in the United States used generative AI in that quarter, which ranked 21st worldwide. The United Arab Emirates led at 70.1%. (Microsoft AI Economy Institute, 2026)
Pew asks all U.S. adults about chatbots, the Fed asks adults ages 18 to 64 about any generative AI product, and Microsoft estimates usage from device telemetry. None of the three measures writing alone, so read each generative AI adoption rate as an upper limit on AI writing assistant use.
How Do Content Marketers Use AI?
AI content marketing, from AI copywriting to long-form drafts, is the most-surveyed corner of AI writing, and three independent surveys land close together on adoption.
- 97% of content marketing programs use AI, up from 90% in 2025 and 83% in 2024. (Siege Media, 2026, 353 marketers)
- 87% of 879 content marketers surveyed by Ahrefs use AI to create or help create content. (Ahrefs, 2025)
- 9 in 10 B2B marketers already use AI to produce content, yet fewer than 4 in 10 say it has improved performance. (Content Marketing Institute, October 2025)
- 74% of content marketers use AI for ideation, 61% for brainstorming and outlining, and 44% for drafting. Outlining fell from 72% and drafting from 57% a year earlier. (Siege Media, 2026)
- 38% use AI for editing, double the 19% recorded in 2025. (Siege Media, 2026)
- 1% say their work is 100% AI-generated. The most common answer, from 48%, is that AI touches 11% to 40% of their work. (Siege Media, 2026)
The task data is where many roundups go wrong. Drafting and outlining use fell in Siege Media’s survey while editing doubled, which suggests marketers are moving AI away from the first draft and toward the cleanup.
How Big Is the AI Writing Market?
No agreed AI writing market size exists. Analyst firms slice the market differently and most full reports sit behind paywalls, so the figures below describe three nested markets, from all AI down to AI content creation.
- $390.9 billion was the size of the global AI market in 2025. It is projected to reach $3.5 trillion by 2033, a 30.6% compound annual growth rate. (Grand View Research)
- $22.21 billion was the size of the generative AI market in 2025, projected to reach $324.68 billion by 2033 at a 40.8% annual rate. (Grand View Research)
- $36.4 billion is the forecast growth of the generative AI in content creation market between 2024 and 2029, a 32.2% annual rate. (Technavio)
- 45% of B2B marketers name AI-powered tools among their top three areas for increased investment in 2026. (Content Marketing Institute, October 2025)
Other roundups quote narrower “AI writing tool” market estimates. We couldn’t trace those to a source we could open, so they aren’t included here.
How Is AI Changing Search and SEO?
AI-generated text is also changing how written content gets found, both in Google’s AI Overviews and in chatbots.
- 60% of U.S. adults say they read the AI summaries at the top of search results. (Pew Research Center, 2026)
- 8% of Google visits ended in a click on a traditional result when an AI summary appeared, compared with 15% when none did. Links inside the summary drew a click on 1% of visits. (Pew Research Center, 2025, browsing data from 900 U.S. adults)
- 0.011 is the correlation between a page’s share of AI content and its Google ranking across 600,000 pages, which is effectively zero. In the same study, 86.5% of top-20 pages contained some AI content. (Ahrefs, 2025)
- 51% of B2B software buyers start their research in an AI chatbot more often than in Google. (G2, 2026, 1,076 buyers)
For anyone publishing AI-assisted content, the ranking data and the click data point in different directions. Google doesn’t appear to penalize AI text, but AI Overviews leave fewer clicks to win.
How Much Time Does AI Writing Save? Productivity and Quality Statistics
Productivity gains are the main argument for AI writing tools, and they’re also where vendor claims run furthest ahead of the evidence. The first statistic below was measured in a controlled experiment. The other five are self-reported, so read them as what users believe AI saves them.
- 40% less time and 18% higher quality: in an experiment with 453 college-educated professionals, those given ChatGPT finished occupation-specific writing tasks 40% faster, and blind evaluators graded their work 18% higher. (Noy and Zhang, Science, 2023)
- 5.4% of work hours is the average time generative AI users said they saved in the previous week, about 2.2 hours in a 40-hour week. (Federal Reserve Bank of St. Louis, November 2024 survey)
- 20.5% of weekly users said AI saved them four hours or more, while 33.0% said it saved an hour or less. (Federal Reserve Bank of St. Louis, November 2024 survey)
- 33% is the Fed’s estimate of how much more productive workers are in each hour they use generative AI. (Federal Reserve Bank of St. Louis, 2025)
- 1.6% of all U.S. work hours were saved once non-users are counted, which implies a labor productivity gain of up to 1.3% since ChatGPT launched. (Federal Reserve Bank of St. Louis, 2025)
- 30% of U.S. adults say chatbots help their productivity and 5% say they hurt it. On creativity the split is closer, at 21% to 11%. (Pew Research Center, 2026)
Weekly time saved by U.S. workers who used generative AI in the previous week. Source: Federal Reserve Bank of St. Louis, November 2024 survey.
The 40% lab result and the 5.4% workplace figure measure different things. The experiment timed a single writing task, while the Fed measured a whole work week in which AI touched only some of the hours.
Speed also isn’t the same as results. As statistic 11 showed, fewer than 4 in 10 B2B marketers say AI has improved their content performance.
How Much Do AI Writing Tools Cost?
Standard individual plans have settled into a narrow price band, which makes cost one of the few areas where the numbers are easy to compare.
- $20 a month is the list price of the standard individual plan on ChatGPT (Plus, $20), Claude (Pro, $20), and Gemini (Google AI Pro, $19.99). Cheaper paid tiers exist, including ChatGPT Go ($8) and Google AI Plus ($4.99). (Vendor list prices, checked October 2026)
- 25% of content marketers spend less than $100 a month on tools that measure their visibility in AI answers, and another 24% spend $100 to $249. (Siege Media, 2026)
- 25% of content marketers now budget under $5,000 a month for content, down from 48% in 2023. Near-universal AI adoption hasn’t shrunk content budgets. (Siege Media, 2026)
| Tool and plan | List price | Free plan |
|---|---|---|
| ChatGPT Plus | $20 per month | Yes |
| Claude Pro | $20 per month, or $17 billed annually | Yes |
| Google AI Pro (Gemini) | $19.99 per month | Yes |
| Grammarly Pro | $30 per month, or $12 billed annually | Yes |
| Jasper Pro | $69 per seat per month, or $59 billed annually | No, free trial only |
Vendors change plans often, so confirm the current price on each official page before you buy. The gap between monthly and annual billing is worth a look too: Grammarly Pro costs less than half as much per month on an annual plan.
Not sure which plan fits your workload? Our guide on how to choose the right AI tool for your business can help you narrow it down.
What Are the Pros and Cons of AI Writing?
The statistics so far point both ways. This breakdown sorts them into what the evidence supports and where it shows limits, with the statistic number so you can check each claim against its source above.
- Faster drafts. Professionals finished a writing task 40% faster with ChatGPT (statistic 23).
- Better routine writing. The same work was graded 18% higher by blind evaluators (statistic 23).
- Time back each week. Users report saving about 2.2 hours in a 40-hour week (statistic 24).
- No search penalty. AI content share and Google ranking show a 0.011 correlation (statistic 21).
- Low cost to start. Standard plans on ChatGPT, Claude, and Gemini cost about $20 a month, and cheaper tiers exist (statistic 29).
- Results don’t follow automatically. Fewer than 4 in 10 B2B marketers say AI improved performance (statistic 11).
- Gains are uneven. A third of weekly users save an hour or less (statistic 25).
- Fewer clicks for what you publish. Clicks fall from 15% to 8% of visits when an AI summary appears (statistic 20).
- A creativity cost for some. 11% of U.S. adults say chatbots hurt their creativity (statistic 28).
One number sits between the columns. Only 1% of content marketers say their work is fully AI-generated (statistic 14), so nearly all of them still do part of the writing or editing themselves.
Which AI Writing Tools Are Most Popular?
ChatGPT usage leads every survey, but the ranking underneath it changes depending on who you ask. The general public and professional content marketers favor different tools.
- 44% of U.S. adults use ChatGPT, up from 34% in 2025 and 18% in 2023. (Pew Research Center, 2026)
- 24% of U.S. adults use Gemini, 17% use Microsoft Copilot, 14% use Meta AI, 8% use Grok, and 6% use Claude. (Pew Research Center, 2026)
- 80% of content marketers name ChatGPT among the most trustworthy AI tools for their work. Claude follows at 55%, up from 28% a year earlier, and Gemini at 44%, up from 16%. (Siege Media, 2026)
- 62% is ChatGPT’s share as the most-used chatbot for B2B software research, up from 47% a year earlier. (G2, 2026)
| Tool | U.S. adults who use it (Pew, 2026) | Content marketers who rate it most trustworthy (Siege Media, 2026) |
|---|---|---|
| ChatGPT | 44% | 80% |
| Gemini | 24% | 44% |
| Microsoft Copilot | 17% | Not reported |
| Claude | 6% | 55% |
| Perplexity | Not asked | 38% |
Claude shows the widest gap. It ranks sixth among the general public, behind Meta AI and Grok, which this table leaves out, and second among people who write for a living. If you’re choosing a tool for professional writing, the marketer column is the closer match to your situation.
Neither column is market share by revenue, which no vendor publishes for writing use alone. They are the two best public measures of which tools people reach for.
Who Uses AI Writing, and Who’s Holding Back?
Adoption looks very different from one group to the next. These six statistics show where AI writing is already routine and where people are still cautious.
Small Businesses and Industries
- 66% of U.S. small businesses use AI in 2026, up from 58% in 2025 and 23% in 2023. (U.S. Chamber of Commerce, 2026)
- 14.0% of work hours in information services involve generative AI, the highest of any industry. Leisure and accommodation is lowest at 2.3%. (Federal Reserve Bank of St. Louis, 2024 surveys)
If you run a small business, you’re now in the minority if you don’t use AI at all. How much it helps depends heavily on how much of your work is writing and analysis.
Students
- 94% of U.K. undergraduates use generative AI to help with assessed work. The share who paste AI-generated text directly into that work is 12%, up from 3% in 2024. (HEPI, 2026, 1,054 students)
- 54% of U.S. teens ages 13 to 17 have used chatbots for help with schoolwork, and 10% do all or most of their schoolwork that way. (Pew Research Center, 2026)
Both surveys show the same pattern as the marketing data: wide use for support, with a much smaller group handing over the writing itself. Students should check their institution’s rules first, since policies on AI in education still vary by course.
Writers and the Job Market
- 59% of Americans expect AI to lead to fewer jobs for journalists over the next 20 years, and only 5% expect more. (Pew Research Center, 2025, survey conducted August 2024)
That figure is an expectation, not a measured job market impact. We looked for a recent, traceable count of writing jobs lost to AI and didn’t find one we could verify, so none appears here.
Who’s Holding Back
- 27.5% of working-age people in the Global North used generative AI in early 2026, compared with 15.4% in the Global South, and the gap is widening. (Microsoft AI Economy Institute, 2026)
Age matters as well. As statistic 6 showed, adults 50 and older are half as likely as younger adults to use ChatGPT.
How to Use AI Writing Statistics Responsibly
A statistic is only as good as the way you cite it. These five steps take a few minutes and keep a weak number out of your report or pitch.
Open the original source. Follow the link to the organization that ran the study, and confirm the number appears there in the form you plan to quote.
Check what was measured. Decide whether the figure describes adoption, intensity, or saturation, and whether it was measured directly or self-reported in a survey.
Record the year and the sample. Write down when the data was collected and who was asked. A 2023 survey of 300 marketers and a 2026 poll of 5,000 adults support very different claims.
Show disagreement instead of hiding it. When two credible sources give different numbers, quote both and explain what each one counted.
Check again before you reuse it. Sources revise their data. The St. Louis Fed raised its own 2024 adoption estimate from 39.4% to 44.6% a year after publishing it.
How Much Online Content Is AI-Generated, and Can It Be Detected?
Estimates of the AI-generated content percentage of the web disagree sharply. The figures below range from 10% to 90%, and each one is defensible once you see what it counted.
| Source | What it counted | Result |
|---|---|---|
| Europol, 2022 | An expert forecast for 2026 covering all synthetic media, including images and video | As much as 90% |
| Ahrefs, April 2025 | 900,000 newly published English-language pages containing any detectable AI text | 74.2% |
| Graphite, May 2026 | 55,400 articles, counted as AI when AI-generated text outweighs human-written text, averaged across three detectors | About 50% of new articles, stable for five quarters |
| Pew Research Center, July 2026 | 10,000 random English-language pages of any age showing significant signs of AI authorship | 10% |
- 74.2% of newly published English-language pages contained some AI-generated text, but only 2.5% were classified as fully AI-written and 71.7% mixed AI with human writing. (Ahrefs, April 2025)
- 10% of English-language web pages in a random sample showed significant signs of AI authorship. Among pages published after ChatGPT launched, the share rose to more than one-third. (Pew Research Center, July 2026)
- 61.22% was the average false-positive rate when seven AI detectors judged 91 essays written by non-native English speakers, meaning most of that human writing was labeled as AI. (Liang et al., Patterns, 2023)
- 66% of Americans are extremely or very concerned about people getting inaccurate information from AI. (Pew Research Center, 2025, survey conducted August 2024)
The spread comes from three choices. A study can count pages with any AI text or only pages that are mostly AI, it can sample new pages or the whole web, and it has to rely on AI content detection software that makes mistakes.
Three of the four figures in the table depend on a detector (Europol’s is a forecast), and statistic 44 shows how badly AI detector false positives can skew results for some writers. Treat any single figure for human vs. AI content as an estimate, and never treat a detector score as proof about one person’s work.
The Bottom Line on AI Writing in 2026
If you’re deciding whether to adopt an AI writing tool, the evidence favors using one for drafts and edits while a person stays responsible for accuracy. Plan on saving a couple of hours a week, not half your workload. Some people should expect less: a third of weekly users save an hour or less (statistic 25). Results depend on what you do with the time.
If you’re here to cite these numbers, link the original study and say what it measured. Every statistic on this page was checked against its source in October 2026, and the methodology note below explains how.
Sourcing and Methodology
Each statistic on this page links to the organization that produced it. We left out any figure we couldn’t trace to a named study or dataset, including several that appear on most competing lists. Survey results describe what respondents reported, not what was independently measured, and we say so where it matters. Where credible sources disagree, we show more than one figure and explain what each counted. Tool prices are vendor list prices, and no tool was hands-on tested for this article.
This page was last reviewed in October 2026. If you spot a newer figure or a broken source link, tell us and we’ll update it. For a wider view of adoption, see our AI tools statistics for U.S. businesses.
Frequently Asked Questions About AI Writing Statistics
What percentage of online content is AI-generated?
It depends on what you count. Pew Research Center found that 10% of English-language web pages in a July 2026 sample showed significant signs of AI authorship, rising to more than one-third of pages published after ChatGPT launched. Ahrefs found that 74.2% of newly published pages contained at least some AI text, though only 2.5% were fully AI-written.
How many people use AI writing tools?
About half of U.S. adults use AI chatbots. Pew Research Center put the figure at 49% in 2026, up from 33% in 2024, with 24% using one every day. Among content marketers the share is far higher: 97% say their program uses AI, according to Siege Media’s 2026 survey of 353 marketers.
Will AI replace writers?
The current data shows AI changing writing work more than eliminating it. Only 1% of content marketers publish fully AI-generated work, and use of AI for drafting fell from 57% to 44% in Siege Media’s 2026 survey. Public expectations are gloomier: 59% of Americans told Pew Research Center they expect fewer journalism jobs over the next 20 years.
Can AI writing be detected?
Not reliably. In a Stanford study published in the journal Patterns, seven AI detectors wrongly labeled about 61% of essays by non-native English writers as AI-generated. Detectors can help estimate AI use across thousands of pages, but a score on a single document isn’t proof, and it shouldn’t be the only basis for an accusation.
Does AI-generated content hurt Google rankings?
No, according to Ahrefs’ analysis of 600,000 pages. It found a 0.011 correlation between the share of AI content on a page and its Google ranking, which is effectively zero. Google’s spam policies target low-value content produced at scale, whatever tool made it, so quality matters more than how the first draft was written.
Do AI Overviews reduce website traffic?
Yes, according to Pew Research Center. It tracked the browsing of 900 U.S. adults in March 2025 and found people clicked a traditional search result on 8% of visits when an AI summary appeared, compared with 15% when none did. Links inside the summary itself were clicked on about 1% of visits.
What is the 30% rule in AI?
There’s no official 30% rule for AI. The phrase is used loosely online to mean that people should keep a meaningful share of the work, often put at 30%, for review and judgment. We couldn’t trace it to a study or a standard. Google sets no percentage threshold and evaluates content on quality.
References
-
Pew Research Center
“How Much of the Internet Is Written With AI?”
-
Pew Research Center
“Americans and AI 2026: Chatbots, Smart Devices and Views on Impact”
-
Siege Media
“The 7 Content Marketing Trends Shaping 2026”
-
Federal Reserve Bank of St. Louis
“The State of Generative AI Adoption in 2025”
-
Ahrefs
“74% of New Webpages Include AI Content (Study of 900k Pages)”
-
Federal Reserve Bank of St. Louis
“The Impact of Generative AI on Work Productivity”
-
Science
“Experimental Evidence on the Productivity Effects of Generative Artificial Intelligence”
I’m Shamim Sarker, founder of SoftwareAdvisorHub.com — an independent software review platform built after 8+ years of hands-on SaaS testing. I never accept payment for positive reviews, and I always disclose affiliate relationships upfront.
My goal: Help you find the right software without wasting money on the wrong one.
Areas of Expertise: SEO Tools · Email Marketing · AI Writing Tools · Ecommerce Platforms · CRM Software · Project Management · Analytics
Every review is based on real hands-on testing — no paid placements, no vendor influence. Affiliate links are always disclosed.
