In 2025, 86% of global creators reported using generative AI, on a sample of 16,000 across eight countries (Adobe Creators’ Toolkit). In the same window, 52% of creators reported career burnout and 37% said they were considering leaving the profession (Billion Dollar Boy / Censuswide, n=1,000 US+UK creators, July 2025). The popular framing of these two numbers, that AI is either rescuing creators or replacing them, fits neither dataset. The picture that does fit, once five other large samples are layered in, is more useful and more boring: creators with structured workflows around AI are reporting growth; creators without that scaffolding are reporting strain. This article walks the data, names the scaffolding gap, and flags the caveats.
Methodology and Sources
Every figure cited below appears in one of the following datasets. We list n-size, fielding window, and source URL up front so a reader can audit any claim against the underlying study.
Creator AI Adoption Is Broad, Not Niche
Five independent samples landed in similar territory in 2025. Adobe’s Creators’ Toolkit reported 86% of global creators using generative AI (n=16,000). Upwork’s Future Workforce Index put freelancer AI use at 84%, about 2.2 times the rate among full-time employees, with the same share (84%) reporting they were excited about AI reshaping their workflow. On the outcome side, 76% of creators in Adobe’s sample said AI helped grow their business or personal brand, and 81% said it enabled them to make content they could not have produced otherwise. Platform telemetry told the same story from a different angle: YouTube CEO Neal Mohan disclosed in his January 2026 annual letter that more than one million YouTube channels used AI creation tools every single day in December 2025.
The adoption snapshot
| Signal | Share | Source | Sample |
|---|---|---|---|
| Use generative AI | 86% | Adobe Creators' Toolkit 2025 | n=16,000, 8 countries |
| Freelancers regularly using AI | 84% | Upwork Future Workforce Index | 2.2x full-time employees |
| Excited about AI reshaping workflow | 84% | Upwork Future Workforce Index | Skilled freelancers |
| Enabled work otherwise impossible | 81% | Adobe Creators' Toolkit 2025 | n=16,000 |
| AI grew business or personal brand | 76% | Adobe Creators' Toolkit 2025 | n=16,000 |
Section takeaway. Across five large 2025 samples, creator AI adoption clusters between 76% and 86%, with similar majorities reporting positive business outcomes. Adoption is no longer an early-adopter signal.
Adoption Doubled in Six Months
The level is high. The slope is steeper. URLgenius’s Creator Trend Index reported creator weekly AI use jumping from 19% to 44% over a six-month window, April to October 2025, with overall creator AI adoption up 131% year over year. The URLgenius sample is small (n=215) and should be treated as directional, but the slope is corroborated by Hootsuite’s larger Social Media Trends 2025 study (n=3,864 across 99 countries), which reported 180% year-over-year growth in social marketers using AI for image editing and 95% YoY growth in using AI to rewrite text outright.
Six-month and YoY acceleration
| Signal | Change | Source | Sample |
|---|---|---|---|
| Social marketers using AI for image editing | +180% YoY | Hootsuite Social Media Trends 2025 | n=3,864, 99 countries |
| Creator AI adoption | +131% YoY | URLgenius Creator Trend Index | n=215 US creators |
| Creator weekly AI use (6 months) | 19% to 44% | URLgenius Creator Trend Index | Apr 2025 to Oct 2025 |
| Using AI to rewrite text | +95% YoY | Hootsuite Social Media Trends 2025 | n=3,864 |
Section takeaway. Creator AI use is not just high in 2025. It is roughly doubling on multiple measures within twelve months, with the steepest slope on visual-editing use cases.
Underneath the Adoption Curve, a Burnout Curve
The same year that delivered 86% adoption delivered the worst burnout signals the creator economy has on record. Billion Dollar Boy / Censuswide (n=1,000 US and UK creators, July 2025) reported 52% of creators experiencing burnout in their career and 37% considering leaving the profession. Workforce-wide, ManpowerGroup’s Global Talent Barometer 2026 (n=13,918 workers across 19 countries) put worker burnout at 63% and the share fearing AI may replace their job within two years at 43%, up five percentage points YoY. On the enterprise side, S&P Global Market Intelligence found that the share of companies scrapping the majority of their AI initiatives jumped from 17% to 42% in a single year.
The two curves, adoption and strain, sit on the same dataset year. They do not cancel each other out. They describe two populations: one that has integrated AI into a workflow and is reporting growth, and one that is using AI under stress and is reporting fatigue.
Adoption vs strain, same year
| Signal | Share | Source | Population |
|---|---|---|---|
| Creators using generative AI | 86% | Adobe Creators' Toolkit | Emerging + semi-pro creators, n=16,000 |
| Creators experiencing burnout | 52% | Billion Dollar Boy / Censuswide | US+UK creators, n=1,000 |
| Creators considering leaving profession | 37% | Billion Dollar Boy / Censuswide | US+UK creators, n=1,000 |
| Workers fearing AI will replace their job in 2 yrs | 43% | ManpowerGroup Global Talent Barometer 2026 | Workforce-wide, n=13,918 |
| Companies scrapping most AI initiatives | 17% to 42% | S&P Global Market Intelligence | Enterprise survey, 2024 vs 2025 |
Section takeaway. Record creator AI adoption (86%) and record creator burnout (52%) are the same year’s numbers. The data describes coexistence, not contradiction.
The Scaffolding Gap
If the same year produced 86% adoption and 52% burnout, the question is what predicts which side of the line a creator lands on. The data points to one factor more than any other: whether the user has structured support around the tool. Three signals from ManpowerGroup’s n=13,918 sample land in the same place. 56% of the global workforce received no workplace AI training in the past year. 57% reported no access to mentorship. Worker confidence in using technology dropped 18 percentage points year over year, with the sharpest fall among Baby Boomers (-35 percentage points).
A fourth signal, from the buying side, hits the same theme from a different angle. Hootsuite / Censuswide (n=500 social media managers + n=500 marketing leaders, June 2025) found that 81% of marketers admit their AI budgets are being wasted on tools not fit for purpose. The deployment is happening. The structure around the deployment is not.
Training, mentorship, and confidence
| Signal | Share / change | Source | Sample |
|---|---|---|---|
| AI budgets wasted on tools not fit for purpose | 81% | Hootsuite / Censuswide | n=500 + n=500, marketers |
| Workforce with no access to mentorship | 57% | ManpowerGroup | n=13,918, 19 countries |
| Workforce with no AI training, past year | 56% | ManpowerGroup | n=13,918, 19 countries |
| Drop in worker tech confidence, year over year | -18% YoY | ManpowerGroup | n=13,918, 19 countries |
Section takeaway. Roughly half the workforce got no training, no mentor, and lost confidence year over year. The companion dataset on marketing buyers shows 81% admitting their AI budgets are wasted on the wrong tools. The pattern is consistent: deployment is moving faster than the structure around it.
What Separates Thrivers from Burnouts
If the diagnosis is a scaffolding gap, the next data question is what scaffolding actually looks like in practice. Anthropic’s Economic Index, published September 2025, gives one signal. Across sampled Claude.ai consumer conversations, augmentation (collaborative use with a human in the loop) edged ahead of automation, 52% to 45%. Earlier reports had automation leading. The flip is small and limited to one product’s traffic, but the direction matters: when users hold the wheel, the work tends to land better.
Augmentation as a habit takes structure to maintain. The creators reporting growth in the Adobe and Upwork samples are not freelance writers pasting “make this better” into a chat box at midnight. They are running the tool inside a workflow that catches its failure modes, edits its drafts, and forces specificity where the model wants to default to a generic answer. The creators reporting burnout in the Billion Dollar Boy sample are operating under financial pressure with limited training and limited mentor access, exactly the conditions the ManpowerGroup data flags.
Section takeaway. Augmentation, not access, predicts which creators are reporting growth. Structured workflows (named rules, reusable templates, identity-lock, failure-mode catches) are the operational form of “training” that the workforce-wide data shows is missing for the majority.
The Trust Gap, Audience Side
The scaffolding gap also shows up on the consumer side, which is the side that pays creators. Meltwater and YouGov’s “Trust in the Age of GenAI” study (approximately 10,000 consumers across seven markets, April 2026 release) found that 86% of consumers say AI-generated content should be disclosed and 32% would trust a brand less if its content is AI-generated. These are not fringe positions. They are majority and near-third positions on a 10,000-person, 7-country sample.
For a creator with disclosure built into the workflow, this is a non-event. For a creator without it, this is reach and trust risk on every platform that throttles or labels AI content. The cost of operating without scaffolding compounds at the platform layer, where Pinterest, TikTok, Meta, YouTube, and Etsy have all rolled out AI-content policies during 2024-2025.
Section takeaway. 86% of consumers want disclosure; 32% will trust a brand less for AI content. Both numbers raise the cost of running AI workflows without structured disclosure habits, especially on platforms with active labeling.
Caveats and Counter-Balance
Three caveats are owed to anyone citing this article.
Sample mismatch on the headline number. Adobe’s 86% is drawn from a sample of emerging and semi-professional creators, not full-time creative-industry professionals. The figure is defensible because Adobe’s methodology is published, but it should not be cited as “86% of all creative pros use AI.” The Upwork freelancer figure (84%) and URLgenius creator figure both corroborate the directional signal on different populations, which is the safer way to cite the trend.
Population mismatch on the burnout-fear numbers. ManpowerGroup’s 63% burnout and 43% AI-replacement-fear are workforce-wide across 19 countries, not creator-only. The Billion Dollar Boy figures (52% creator burnout, 37% considering leaving) are creator-specific and smaller (n=1,000). Use the right number for the right claim.
Small-sample signal on the acceleration trend. URLgenius’s six-month doubling and 131% YoY are based on n=215 US creators. Hootsuite’s image-edit and rewrite YoY figures are on n=3,864 across 99 countries, and they corroborate the slope. The acceleration trend is real but the precise magnitude varies by sample.
The counter-balance worth holding is the Anthropic augmentation finding: on the product where users opted into human-in-the-loop use, augmentation moved ahead of automation in 2025. The frame of “AI replaces creators” is not the frame the actual usage data is pointing toward. The usage data is pointing toward “AI rewards users with structure and punishes users without it.” That is a meaningfully different story.
What This Means for Operators
For anyone running creator workflows, the actionable read of this dataset is straightforward. Adoption is settled; you do not need to debate whether to use AI. Acceleration is steep; the workflows that worked six months ago are already behind. Scaffolding is the variable that predicts outcomes. Written rules, reusable templates, identity-lock conventions, disclosure habits, and failure-mode catches are the things that separate the 76% who are growing from the 37% who are considering quitting. None of it is glamorous. All of it is testable against the data above.
This article exists partly because the team behind it built a 125-prompt image library governed by a 19-rule production methodology, as one explicit attempt to make the scaffolding the data shows is missing. Every prompt carries the same rules. The rules are written down. The library is the operational form of the argument.
Key Takeaways
- Adoption is settled. 86% of global creators use generative AI (Adobe, n=16,000); 84% of freelancers use it regularly (Upwork). The directional signal is consistent across five large 2025 samples.
- Adoption is still accelerating. Creator weekly AI use jumped from 19% to 44% in six months (URLgenius); social marketers using AI for image editing grew 180% YoY (Hootsuite).
- Burnout is at the same record level adoption is. 52% creator burnout, 37% considering leaving (Billion Dollar Boy); 63% workforce burnout and 43% fearing AI replacement within two years (ManpowerGroup).
- The scaffolding gap is the most consistent signal. 56% of the workforce got no AI training, 57% no mentorship, worker tech confidence dropped 18 points YoY (ManpowerGroup); 81% of marketers admit AI budgets are wasted on the wrong tools (Hootsuite).
- Augmentation is winning where it has been measured. Anthropic’s September 2025 Economic Index reported 52% augmentation vs 45% automation on Claude.ai consumer conversations.
- Audience trust raises the cost of running unscaffolded. 86% of consumers want AI content disclosed; 32% will trust a brand less for using it (Meltwater × YouGov, n~10,000).
Frequently Asked Questions
Q: How many creators use generative AI in 2026?
A: About 86% of global creators reported using generative AI in Adobe’s Creators’ Toolkit 2025 survey (Harris Poll, n=16,000 across eight countries, fielded September 2025). The Adobe sample is emerging and semi-professional creators, not full-time creative-industry professionals. A separate Upwork Future Workforce Index put freelancer use at 84%, roughly 2.2 times the rate of full-time employees. URLgenius tracked creator weekly AI use jumping from 19% to 44% across a six-month window. The directional signal is consistent across five samples.
Q: How much of the creator workforce is burned out?
A: 52% of creators reported experiencing burnout in their career, and 37% said they were considering leaving the profession, according to Billion Dollar Boy and Censuswide (n=1,000 US and UK creators, July 2025). The top severity driver was financial instability (55%) followed by creative fatigue (40%). Workforce-wide, ManpowerGroup’s Global Talent Barometer 2026 (n=13,918, 19 countries) reported 63% of workers experiencing burnout driven by stress and workload.
Q: Is AI making creator burnout worse or better?
A: The data does not support either single direction. 76% of creators in Adobe’s sample said AI helped them grow their business and 81% said it enabled work otherwise impossible. In the same year, 43% of workers in ManpowerGroup’s data feared AI would replace their job within two years (+5 percentage points year over year) and the share of companies scrapping the majority of their AI initiatives jumped from 17% to 42% in one year (S&P Global Market Intelligence). The pattern across studies is that outcomes vary sharply by whether the user has structured workflows, training, and mentorship around the tool.
Q: What share of workers have received AI training at work?
A: 44%. ManpowerGroup’s Global Talent Barometer 2026 found that 56% of the global workforce received no workplace AI or technology training in the past year, and 57% reported no access to mentorship. In the same dataset, worker confidence in using technology fell 18 percentage points year over year, with the sharpest drop among Baby Boomers (-35%).
Q: Does the data say AI is automating creators or augmenting them?
A: As of September 2025, augmentation pulled slightly ahead of automation on Claude.ai consumer conversations. Anthropic’s Economic Index reported 52% of sampled conversations involved collaborative augmentation, vs 45% involving automation (the remainder unclassified). Earlier reports had automation leading. The flip is small and one product’s data, but the direction is meaningful for any “human-in-the-loop is winning” framing.
Sources
- Adobe Creators’ Toolkit 2025 (Harris Poll, n=16,000): https://news.adobe.com/news/2025/10/adobe-max-2025-creators-survey
- Anthropic Economic Index, September 2025 report: https://www.anthropic.com/research/anthropic-economic-index-september-2025-report
- Billion Dollar Boy / Censuswide, “Over half of creators face burnout,” July 2025: https://www.billiondollarboy.com/news/over-half-of-creators-face-burnout/
- Fortune, “AI companies, employee fatigue, failure,” June 2025 (citing S&P Global Market Intelligence): https://fortune.com/2025/06/11/ai-companies-employee-fatigue-failure/
- Fortune, “AI workers toxic relationship, trust confidence collapses, training, Manpower Group,” January 2026: https://fortune.com/2026/01/21/ai-workers-toxic-relationship-trust-confidence-collapses-training-manpower-group/
- Hootsuite Social Media Trends 2025 and Hootsuite / Censuswide marketer survey: https://www.hootsuite.com/newsroom/press-releases/hidden-cost-of-outdated-ai-revealed-as-it-drains-marketers-time-budgets
- ManpowerGroup Global Talent Barometer 2026: https://www.manpowergroup.com/en/news-releases/news/global-talent-barometer-2026-ai-use-accelerates-as-worker-confidence-falls-and-job-hugging-takes-hold
- Meltwater × YouGov “Trust in the Age of GenAI” April 2026: https://www.meltwater.com/en/blog/generative-ai-trust-yougov
- Outlierkit summary of YouTube CEO Neal Mohan annual letter, January 2026: https://outlierkit.com/resources/youtube-ai-tools-1-million-channels-2026/
- Slashdot summary of Upwork Future Workforce Index Q2 2025: https://slashdot.org/story/25/06/30/1959225/freelancers-using-ai-tools-earn-40-more-per-hour-than-peers-study-says
- URLgenius Creator Trend Index Q4 2025 (via BusinessWire): https://www.businesswire.com/news/home/20251016686832/en/Commerce-Omnifluencers-Drive-131-Rise-in-AI-Adoption-Now-Core-to-Creator-Weekly-Workflow-Earnings