136 Generative AI in Marketing Statistics October 2026

Generative AI in marketing statistics for 2026 show a job that has been rebuilt around the tools faster than anyone has measured what the tools return. 96% of the world's largest brand owners now use generative or agentic AI in their marketing, up from 45% in 2023 (WFA, Sep 2026). Yet 76% of marketing leaders spend three hours or more a week fixing what it produces (Optimizely, Jun 2026).

This page follows the marketer's side of the technology, from the tasks AI took over and what teams pay for it to what experiments say it returns, what it does to jobs, how audiences react and where brands carry risk. Every number links to the organisation that ran the survey, experiment or filing, and each was checked on the live source. Where a vendor ran the survey, the citation names it so you can weigh it. Broader AI figures sit on our generative AI statistics page, a general view of marketers' AI use on AI marketing statistics, and the search side on AI search statistics.

What marketing work do teams now hand to AI?

Writing came first and still leads. The newer shift is into research, media buying and agents, where a minority of teams have moved and most are still testing.

Who in marketing uses AI, and how often?

Nearly all of them. Surveys of brand owners, agencies, content teams and PR professionals each find at least three in four using AI.

• 97% of 1,415 marketing leaders at companies with 500 or more employees use AI in their daily creative work (Canva, May 2026).
• 93% of 301 marketers surveyed use AI to create or help create content, up from 87% a year earlier (Ahrefs, Sep 2026).
• Nine in 10 US marketing agencies use generative AI (Forrester, Jun 2026).
• 76% of 564 PR professionals use generative AI, a share unchanged from the year before (Muck Rack, Jan 2026).
• In a McKinsey survey reproduced in the Stanford AI Index, 51% of respondents in consumer goods and retail reported AI use in marketing and sales (Stanford HAI, Apr 2026).
• 41% of marketing leaders describe AI as a director on their team, and 39% call it a collaborator (Canva, May 2026).

Which tasks get done with AI first?

Content comes first, research and ideation next. Media buying and strategy trail by 25 points or more.

• 80% of top brand owners use AI for content creation, 74% for ideation and production, and 72% for research and insights (WFA, Sep 2026).
• 54% use it for media buying and optimisation and 48% for strategy and planning (WFA, Sep 2026).
• 74% of PR professionals use AI for writing and content creation and 68% for strategy and planning, while 30% use it for media lists and outreach (Muck Rack, Jan 2026).
• 74% of US agencies use generative AI to summarise documents and communications, and 70% apply it to research and competitive intelligence (Forrester, Jun 2026).
• Use of AI for image creation rose from 19% to 43% of surveyed marketers in a year (Ahrefs, Sep 2026).
• 60% of marketing leaders use AI to support personalisation across channels, yet nearly half lack the tools to coordinate those experiences (Braze, Feb 2026).

Where does the move from assistant to agent stand?

Mostly still at assistant. Agents run real work at a minority of teams, and the people running them want a human sign-off.

• 42% of CMOs use generative AI only as an assistant for individual tasks in a handful of workflows (BCG, Jun 2026).
• Just under a third of CMOs say they have transformed significant parts of their function with agents (BCG, Jun 2026).
• 46% of top brand owners use AI for workflow automation or autonomous agents (WFA, Sep 2026).
• Half of US agencies use agentic AI for marketing execution (Forrester, Jun 2026).
• Only 12% of PR professionals who use AI say they use AI agents, and 89% want clear human approval before an agent publishes or sends anything (Muck Rack, Jan 2026).
• 19% of consumers use AI agents to deal with brands today, and Braze expects that to reach 46% by the end of 2026 (Braze, Feb 2026).

How much marketing material does AI now turn out?

Output is where the effect is easiest to see. Few teams publish raw model text, but the share of marketing material with a model in the chain keeps rising.

How much written copy comes from language models?

A large and growing share, almost always with a person editing it.

• 83.5% of marketers say they are expected to produce more content because of AI, and 35.7% say much more (HubSpot, Apr 2026).
• 69% of marketers say the share of AI content they publish grew over the past year, rising to 74% among AI users (Ahrefs, Sep 2026).
• 7% of AI-using marketers say more than 75% of what they publish is unedited AI output, up from 4% in 2025, while 40% publish none (Ahrefs, Sep 2026).
• Klarna's in-house Copy Assistant lets it use AI for 80% of all copywriting (Klarna, May 2024).
• 45% of PR professionals say they must edit most of the text AI gives them, down from 61% in 2024 (Muck Rack, Jan 2026).
• About 18% of financial consumer complaint text looked model-assisted by late 2024, as did just under 10% of job postings from small firms (Liang et al., arXiv, Feb 2025).

How many images, videos and ads do models generate?

Counts are rarely published, and the ones that are show production at platform scale.

• Klarna generated more than 1,000 images with AI in the first quarter of 2024 and cut its image development cycle from six weeks to seven days (Klarna, May 2024).
• Posts referencing AI video tripled between March 2025 and February 2026, and their engagement rose 557% (Meltwater and YouGov, Apr 2026).
• Ads from the Ju and Aral experiment, which produced 11,024 of them, were tested in a field experiment on X that drew about 5 million impressions, where better text lifted click-through and better images lowered cost per click (Ju and Aral, arXiv, Feb 2026).
• Coca-Cola's 2025 holiday ad was reportedly assembled from 70,000 AI-generated clips (System1, Nov 2025).
• One ad platform dataset used to compare AI and human visuals spans 16 billion impressions and 116 million clicks (Marketing Science Institute, Feb 2025).

How much are marketing teams paying for AI?

Spending follows use. At the top of the market, budgets are growing quickly, and by one account sales and marketing take half of all generative AI money.

How large are AI budgets in marketing?

Large and rising, with marketing now steering more of the decisions.

• 99% of marketing leaders plan to increase AI investment in 2026 (Canva, May 2026).
• Martech and data is now the top AI investment area for CMOs, up 11 to 12 percentage points since 2025 (BCG, Jun 2026).
• Roughly half of CMOs say marketing now leads AI investment decisions in its function, while 72% of CEOs see themselves as the main AI decision maker across the enterprise (BCG, Jun 2026).
• MIT NANDA's report says 50% of generative AI budgets go to sales and marketing, though another passage puts the share near 70% in its hypothetical allocation exercise (MIT NANDA, Jul 2025).
• 75% of PR professionals pay for an AI service, up from 57% a year earlier (Muck Rack, Jan 2026).
• 61% of US agencies still treat AI as a cost of business with limited direct monetisation, and 31% plan to monetise agentic AI within 24 months (Forrester, Jun 2026).

What does the next AI budget buy?

Orchestration, creative tooling and visibility inside AI assistants. The last is the least mature.

• 41% of the highest-investing CMOs are dramatically increasing spend on end-to-end workflow orchestration (BCG, Jun 2026).
• Adobe's AI-first annual recurring revenue grew more than 150% year over year in its third fiscal quarter (Adobe, Sep 2026).
• Adobe's Creative and Marketing Professionals customer group generated $4.65 billion of subscription revenue in the quarter, up 13% (Adobe, Sep 2026).
• A net 75% of marketers plan to raise investment in visibility inside AI assistants in 2027, while only 24% of consumers trust brand recommendations from those assistants (Kantar, Oct 2026).
• 73% of top brand owners monitor and measure their AI visibility, yet only 8% have a clear strategy for it (WFA, Sep 2026).
• Forrester predicts leading advertisers will cut display ad budgets by 30% as consumers move to AI summaries and chat interfaces (Forrester, Oct 2025).

What do controlled tests say AI gives marketers?

The strongest evidence here comes from controlled experiments. They agree that AI makes marketing work faster and, for text, better, with weaker results for images and for already-skilled people.

How much faster and better is work done with AI?

Faster by roughly a sixth to two fifths, depending on the task, and better on writing.

• In a preregistered experiment with 453 college-educated professionals, ChatGPT cut the time taken on writing tasks by 40% and raised output quality by 18% (Noy and Zhang, Science, Jul 2023).
• Workers exposed to ChatGPT in that experiment were twice as likely to use it in their real job two weeks later (Noy and Zhang, Science, Jul 2023).
• In a field experiment with 776 Procter and Gamble professionals, individuals using AI matched the performance of teams working without it (NBER, Apr 2025).
• Individuals with AI spent 16.4% less time on the task than the control group, and teams with AI spent 12.7% less (HBS AI Institute, Apr 2025).
• Marketing professionals whose employers encourage chatbot use report saving 6.8% of their work hours, and 69.8% say chatbots improve their work quality (Humlum and Vestergaard, University of Chicago BFI, Apr 2025).
• Access to AI story ideas raised the novelty of writers' stories by 5.4% with one idea and 8.1% with five, yet the AI-assisted stories were more similar to one another (Doshi and Hauser, Science Advances, Jul 2024).
• GPT-4 with access to personal data was more persuasive than a human opponent 64.4% of the time in debates where the two were not equally persuasive (Salvi et al., Nature Human Behaviour, May 2025).

Do AI-made ads win more clicks?

Sometimes, and mostly when the audience cannot tell AI made them.

• Human-made ads that looked like AI averaged a 0.55% click-through rate, against 0.67% for human-made ads that did not (Marketing Science Institute, Feb 2025).
• Keyword-specific AI ad copy beat templates in online A/B tests by up to 9% on click-through rate, with 12% more impressions and 0.38% lower cost per click (Liu et al., arXiv, Jun 2025).
• In randomized field tests at a cross-border retailer, generative AI raised sales by up to 16.3% in some workflows, and return rates and ratings did not worsen (Fang et al., arXiv, Jun 2026).
• Amazon says advertisers using its Ads Agent targeting see 8% lower cost per impression and 6% lower cost per acquisition (Amazon, Jul 2026).
• Coca-Cola's AI-made 2025 holiday ad earned System1's maximum 5.9-star rating (System1, Nov 2025).
• In an NIQ test with more than 2,000 viewers and EEG on about 150, AI-generated ads produced weaker memory activation than traditional ads, even when viewers rated them high quality (NIQ, Dec 2024).

What does the cleanup cost?

Hours a week, and most of it lands on the people closest to the output.

• 48% of marketing leaders say fact-checking and hallucination review created more extra work than any other factor, and 40% name time spent moving information between disconnected systems (Optimizely, Jun 2026).
• Only 4% say AI saves them time at every stage of the process, and only 19% work from a single integrated AI platform (Optimizely, Jun 2026).
• 25% admit they frequently or always publish AI content they know is not fully on-brand when deadlines press, and 30% say they frequently or always pass off AI-generated work as their own (Optimizely, Jun 2026).
• 69% of C-suite leaders say AI adoption is fully aligned across their organisation, while 27% of analysts agree (Optimizely, Jun 2026).
• 41% of AI-using marketers say it has not saved them money on writers or freelancers (Ahrefs, Sep 2026).
• 31% of B2C CMOs and 20% of B2B CMOs say their agentic marketing transformation already has a significant, measurable revenue effect (BCG, Jun 2026).
• Two years ago 70% of brand owners named efficiency as their main marketing objective and 30% named effectiveness (WFA, Sep 2026).

How is AI changing marketing careers?

AI skills now carry pay and posting weight. So far the cuts fall on execution roles and entry-level hiring.

Which skills do marketing teams say they lack?

Strategy first, then AI fluency. Teams do not expect tools to close the gap.

• 68.2% of marketers say they understand how to use AI in marketing, up from 47% in 2025 (HubSpot, Apr 2026).
• 72% name strategic thinking as marketing's number one talent shortage, ahead of data strategy (46%) and insight development (36%) (WFA, Sep 2026).
• 56% identify AI fluency as underdeveloped, and 52% rank the inability to upskill existing talent fast enough among their top two talent risks (WFA, Sep 2026).
• Just 1% believe better tools and platforms will be the main differentiator of organisations with strong marketing talent (WFA, Sep 2026).
• 59% of marketers rank AI skills among the top three most critical for future relevance, second only to strategic thinking at 65% (Content Marketing Institute, Feb 2026).
• The average wage premium for AI skills reached 62%, up from 57%, and runs as high as 118% in consumer markets (PwC, Jun 2026).
• 65% of CMOs expect AI to change the CMO role dramatically within two years, while only 32% say significant changes to the CMO skill set are needed (Gartner, Feb 2026).

Which marketing roles are shrinking or growing?

Execution roles are shrinking and strategy roles are holding. Entry-level hiring takes the hit.

• The share of marketing job postings mentioning AI doubled in 2025, while marketing jobs overall remain 27% below pre-pandemic levels (AMA, Jul 2026).
• Influencer marketer roles grew 18% in 2025, the fastest growth of any tracked marketing role (AMA, Jul 2026).
• About 15% of marketing postings on Indeed contain AI-related terms, against roughly 45% in data and analytics and 9% in human resources (Indeed Hiring Lab, Jan 2026).
• Entry-level marketing hiring scored a net minus 19.8 points, while senior hiring scored a net plus 22.3 (Content Marketing Institute, Feb 2026).
• 75% of employed marketers say finding a marketing job is harder than two years ago, up from 54% in 2024, and new roles took 5.2 months to find against 3.1 (Content Marketing Institute, Feb 2026).
• Freelance job posts for automation-prone writing and coding fell 21% within eight months of ChatGPT's launch, and posts for image creation fell 17% after image generators arrived (Demirci, Hannane and Zhu, Management Science, Jan 2025).
• Freelancers in writing-related occupations saw monthly jobs fall 2% and monthly earnings fall 5.2% after ChatGPT's release (Hui, Reshef and Zhou, CESifo, Aug 2023).

Is AI replacing marketing staff?

Not at scale yet. It changes the work first and the headcount later, if at all.

• 10% of marketers say their company has hired staff with AI experience, and 10% say it created specialised AI roles (Content Marketing Institute, Feb 2026).
• 47% say AI is reshaping their team's workflows, roles or tasks extensively or significantly, and only 4% report no impact (Content Marketing Institute, Feb 2026).
• 53% expect AI to augment their role, and only 4% expect it to replace them fully (Content Marketing Institute, Feb 2026).
• 76% of marketers feel they are doing the work of more than one job (Content Marketing Institute, Feb 2026).
• Three quarters of marketing leaders expect creative roles to grow over the next five years (Canva, May 2026).
• 42% of leaders say AI will never fully replicate empathy and emotional intelligence (Canva, May 2026).
• Gartner predicts that by 2027 a lack of AI literacy will rank among the top three reasons CMOs are replaced at large enterprises (Gartner, Feb 2026).

How do audiences react to AI-made marketing?

Audiences accept AI that helps and resist AI that stands in for people. They are also worse at spotting it than they think.

Can people spot AI-made ads and copy?

Half can, and many say they can when they cannot.

• 70% of 3,547 consumers say they can usually spot AI-generated ads because something feels missing (Canva, May 2026).
• In a test with 2,000 UK and US readers, 50% correctly identified the AI-written article (Bynder, Apr 2024).
• In the same test, 56% preferred the AI version when they were not told which was which (Bynder, Apr 2024).
• 72% of consumers say AI makes it difficult to tell what content is truly authentic (Yahoo and Publicis Media, Feb 2024).
• 76% of consumers agree that it is getting to the point where they cannot tell if an image is real (Getty Images, Apr 2024).
• 56% of social media users see AI slop often or very often, and 83% see it at least sometimes (Sprout Social, Feb 2026).
• 13% of social media users know they follow AI-generated influencers and 27% are unsure (Sprout Social, May 2026).

Does AI lower trust and buying intent?

It lowers both for a sizeable minority, and the effect is larger when AI replaces people than when it helps.

• 78% of consumers would rather see ads made by people even if AI could produce better ones, while 68% do not mind AI in ads if it makes them more helpful or relevant (Canva, May 2026).
• 32% would trust brands less if content is AI-generated and 15% would trust them more (Meltwater and YouGov, Apr 2026).
• 52% would become less engaged on suspecting copy is AI-generated, and 26% would find a brand impersonal if its website copy did not feel human-written (Bynder, Apr 2024).
• Acceptance of AI is 53% in entertainment and 47% in advertising, against 28% in influencer content and 21% in news (Meltwater and YouGov, Apr 2026).
• 44% of social media users are uncomfortable with brands using AI influencers, 25% are comfortable and 30% say it depends (Sprout Social, May 2026).
• 50% of Gen Z have unfollowed, muted or blocked a brand or creator because its content felt like AI slop (Sprout Social, Feb 2026).

Do disclosure labels help or hurt?

On the evidence so far they help, and consumers ask for them.

• Ads whose AI disclosure was noticed gained a 47% lift in ad appeal, a 73% lift in ad trustworthiness and a 96% lift in trust in the company (Yahoo and Publicis Media, Feb 2024).
• 86% of consumers say AI-generated content should be disclosed (Meltwater and YouGov, Apr 2026).
• 28% name unlabelled AI-generated content as the one thing they most want brands to stop doing in 2026 (Sprout Social, Feb 2026).
• 70% of consumers believe it will eventually be impossible to tell whether an ad was AI-generated without disclosure (Canva, May 2026).
• 75% of US adults find it acceptable for businesses to use AI for brainstorming or early drafts of ads when that use is disclosed, and 53% accept it for the final text, images or video (Bentley-Gallup, Aug 2026).
• Almost 90% of consumers want to know whether an image was created with AI (Getty Images, Apr 2024).
• Only 22% of AI-using marketers disclose AI use to site visitors, up from 16% in 2025 (Ahrefs, Sep 2026).

What risks do brands take when AI makes the content?

The risks are practical: output nobody reviewed, unclear ownership, and sharing a feed with a flood of low-quality AI material.

How much human review does AI output get?

Most brands require some, and the rules are newer than the tools.

• 66% of marketers trust AI-generated content only if a human reviews it first (Ahrefs, Sep 2026).
• 75% of top brand owners require human oversight of AI-generated outputs, 70% keep approved or prohibited tool lists and 57% require review and approval of outputs (WFA, Sep 2026).
• 55% of top brand owners have a centrally approved list of AI tools with clear governance, up from 18% in 2023 (WFA, Sep 2026).
• 63% of US agencies cite accuracy and bias as a barrier to scaling AI, 62% cite legal concerns and 55% cite privacy and security (Forrester, Jun 2026).
• 51% of PR professionals work somewhere with an AI use policy, up from 21% in 2024, and 43% have been offered AI training (Muck Rack, Jan 2026).
• 65% of marketing leaders say they would pause or adjust their company's AI rollout to strengthen guardrails if given the chance (Optimizely, Jun 2026).

Who owns AI-made assets, and who labels them?

Ownership is narrower than most teams assume, and the labelling duty is moving to platforms and contracts.

• The US Copyright Office concluded that prompts alone do not give users enough human control to make them the authors of AI output (US Copyright Office, Jan 2025).
• The Office has registered hundreds of works that include AI-generated material, covering the human author's contribution (US Copyright Office, Jan 2025).
• TikTok has labelled over 1.3 billion videos as AI-generated, using creator labels, its own detection and C2PA Content Credentials (TikTok, Nov 2025).
• Over 37 million creators have used TikTok's AI-label tool (TikTok, May 2024).
• 59% of top brand owners now add AI clauses to agency contracts covering intellectual property, confidentiality and training data, and 69% require agencies to disclose when AI is used (WFA, Sep 2026).
• Only 13% of brand owners have a great deal of visibility into agency AI use, up from 3% in 2023 (WFA, Sep 2026).

Where does AI-made content hurt brand safety and voice?

Next to low-quality AI material, and in the sameness it creates.

• 61% of media experts are excited about generative AI in digital media, yet 53% say adjacency to AI-generated content will be a top challenge in the coming year (Integral Ad Science and YouGov, Dec 2025).
• Mentions of AI slop in media monitoring data rose ninefold, and 41% of marketing leaders say it is becoming a real challenge (Canva, May 2026).
• 88% of social media users agree that video generation tools have made them trust the news they see on social media less (Sprout Social, Feb 2026).
• Only 30% of marketing leaders believe their brand voice is genuinely unmistakable (Optimizely, Jun 2026).
• Human-AI teams in the Ju and Aral experiment produced more homogeneous, self-similar ads (Ju and Aral, arXiv, Feb 2026).
• Forrester predicts AI-driven privacy breaches will lead to a 20% surge in US consumer class-action lawsuits (Forrester, Oct 2025).
• Forrester also predicts a third of companies will harm customer experience by deploying AI self-service too early (Forrester, Oct 2025).

Methodology and Sources

We included a number only after opening the page that publishes it and confirming the figure and its date. Figures quoted in roundups were followed to the original survey, experiment, filing or dataset, or dropped. Most numbers come from peer-reviewed or working-paper experiments, trade bodies, regulators, company filings and survey firms. Where a software vendor ran the survey (Canva, Optimizely, Braze, Ahrefs, Adobe), the citation names it. Dates are the date of the data or publication shown on the source. Survey samples are not comparable across sources, so the page compares direction and scale rather than exact levels. Forrester items marked as predictions are forecasts. The page is refreshed every month.

Sources

• Adobe, Reports Record Q3 Results (fiscal 2026)
• Ahrefs, 69% of Marketers Are Publishing More AI Content Than Last Year
• Amazon, Q2 earnings: Andy Jassy on what's driving Amazon Ads growth
• American Marketing Association, 2026 State of Marketing Careers Report
• BCG, Mind the Marketing Gap
• Bentley University-Gallup, Americans Aren't Sold on Businesses Using AI in Advertising
• Braze, The 2026 Braze Customer Engagement Review
• Bynder, Study reveals how consumers interact with AI-generated content vs human-made
• Canva with The Harris Poll, The State of Marketing and AI Report 2026
• Content Marketing Institute, 2026 Career and Salary Outlook: Content and Marketing Professionals
• Dell'Acqua et al., The Cybernetic Teammate (NBER Working Paper 33641)
• Demirci, Hannane and Zhu, Who Is AI Replacing? (Management Science)
• Doshi and Hauser, Generative AI enhances individual creativity but reduces the collective diversity of novel content (Science Advances)
• Fang et al., Generative AI and Sales Productivity: Field Experiments in Online Retail (arXiv)
• Forrester and 4As, The State of AI Inside US Marketing Agencies, 2026
• Forrester, 2026 B2C Marketing, CX and Digital Business Predictions
• Gartner, CMO AI blind spot survey
• Getty Images, Building Trust in the Age of AI (VisualGPS)
• Harvard Business School AI Institute, The Cybernetic Teammate
• HubSpot, 2026 State of Marketing
• Hui, Reshef and Zhou, The Short-Term Effects of Generative AI on Employment (CESifo Working Paper 10601)
• Humlum and Vestergaard, Large Language Models, Small Labor Market Effects (University of Chicago BFI)
• Indeed Hiring Lab, January 2026 US Labor Market Update
• Integral Ad Science with YouGov, The 2026 Industry Pulse Report
• Ju and Aral, Collaborating with AI Agents: Field Experiments on Teamwork, Productivity, and Performance (arXiv)
• Kantar, Media Reactions 2026
• Klarna, AI helps Klarna cut marketing agency spend by 25% and run more campaigns
• Liang et al., The Widespread Adoption of Large Language Model-Assisted Writing Across Society (arXiv)
• Liu et al., LLMs for Customized Marketing Content Generation and Evaluation at Scale (arXiv)
• Marketing Science Institute, AI in Disguise (Working Paper 25-146)
• Meltwater and YouGov, Trust in the Age of Generative AI
• MIT NANDA, The GenAI Divide: State of AI in Business 2025
• Muck Rack, State of AI in PR 2026
• NIQ, Research uncovers hidden consumer attitudes toward AI-generated ads
• Noy and Zhang, Experimental evidence on the productivity effects of generative AI (Science)
• Optimizely, 2026 global data study (with Savanta)
• PwC, 2026 Global AI Jobs Barometer
• Salvi et al., On the conversational persuasiveness of GPT-4 (Nature Human Behaviour)
• Sprout Social, Q1 2026 Pulse Survey Analysis
• Sprout Social, Q2 2026 Pulse Survey Analysis
• Stanford HAI, AI Index Report 2026, Economy chapter
• System1, AI or No AI, Coke Gets the Christmas Love
• TikTok, More ways to spot, shape and understand AI-generated content
• TikTok, Partnering with our industry to advance AI transparency and literacy
• US Copyright Office, Copyright and Artificial Intelligence, Part 2: Copyrightability
• WFA and Mediasense, Strategic thinking is the biggest skills gap for marketing
• WFA, 96% of major brands now use AI in marketing (State of AI in Marketing 2026)
• Yahoo and Publicis Media, AI ad disclosure increases consumer trust

Glossary

• AI slop: low-effort, mass-produced AI content, in Sprout Social's wording content generated at scale to increase views and engagement.
• Agentic AI: AI systems that carry out multi-step tasks with limited human input.
• C2PA: Coalition for Content Provenance and Authenticity, the standard behind Content Credentials.
• CMO: chief marketing officer.
• EEG: electroencephalogram, a measure of brain activity.