40+ Statistics on AI Consumer Acceptance in Marketing

What People Really Think About AI Content on Social Media

Generative AI is now woven through the social media feed, and audiences are starting to feel it. The question for brands and agencies is no longer whether people notice AI-generated content, they overwhelmingly do, but whether they accept it. The honest answer from the data is: it depends entirely on what the AI is being used for. People tolerate and even welcome AI in utilitarian roles like shopping and customer service, while rejecting it in creative, cultural and emotional content. Running underneath almost every study is a persistent perception gap: marketers are far more enthusiastic about AI than the consumers they are trying to reach.

Below are 40+ sourced statistics from 2025 and 2026, organised by theme, with the strongest trend lines pulled out first. Figures are flagged by geography (US, UK or global) where the distinction matters. Each is linked to its primary source.

The headline numbers

• Advertisers think audiences like AI ads roughly twice as much as they actually do. 82% of ad executives believe Gen Z and Millennials feel positive about AI ads, but only 45% of those consumers actually do, a 37-point gap that widened from 32 points in 2024. (IAB, Jan 2026)
• Consumer preference for AI-generated creator content collapsed from 60% to 26% in two years. A 34-point fall between 2023 and 2025, even as marketers poured more budget into it. (Billion Dollar Boy, 2025)
• Gen Z want AI nowhere near creative work but don't mind it when shopping. 54% prefer no AI involvement at all in creative work (art, music, books), versus just 13% who feel that way about shopping. (Goldman Sachs, Sept 2025)
• People are largely satisfied with AI customer care but still want to know it's there. 71% were satisfied or very satisfied with AI-supported customer support, yet 83% want brands to disclose when AI is being used. (Emplifi, 2025)

1. Overall sentiment & awareness

• 88.3% of people recognise AI-generated content in their social feeds, and attitudes remain largely negative, it is frequently perceived as “inhuman” and “boring.” (Digital Age in Semiotics & Communication, 2025)
• 50% of US adults are more concerned than excited about AI in daily life; only 10% are more excited than concerned. (Pew Research Center, Sept 2025)
• Across 25 countries, a median of 34% are more concerned than excited about AI; only a median 16% are more excited. The UK falls on the concerned side of the ledger. (Pew Research Center, Oct 2025)
• 53% of US adults say AI will worsen people's ability to think creatively, against just 16% who say it will improve it. (Pew Research Center, Sept 2025)
• Just 35% of US consumers, and 28% in the UK, say they trust AI-generated content. (Emplifi, Apr 2026)

2. Acceptance by use case

This is the single most important pattern in the data. “Consumers reject AI” is simply wrong; “consumers reject AI in the wrong places” is the accurate version.

• The Goldman Sachs split is the clearest single illustration: 54% of Gen Z want no AI in creative work, but only 13% feel that way about shopping, and 38% are comfortable with AI-only results (no human oversight) when shopping, the highest of any category. (Goldman Sachs, Sept 2025)
• On the utility side, sentiment turns positive: roughly 77% feel positive about AI in shopping, 73% use AI tools for brand or product discovery, and 64% believe AI will improve how they engage with brands on social. But 56% still prefer a human for support and only 25% want AI-only. (Emplifi, 2025).
• Consumers are most comfortable with AI in tech, food and entertainment, and most wary in health, finance and personal care. (Emplifi, 2025)
• AI-generated ads are better accepted for innovative, high-tech products and less accepted for traditional or emotional ones, where the AI label “acts like a warning signal.” (NIM, 2025)
• 47% are likely to use generative AI to research purchases and 54% are ready to engage with AI shopping assistants, utility adoption is real and growing. (Attest, 2025)

3. Generational differences

• The counterintuitive headline: Gen Z is the most hostile, not the most receptive. Gen Z consumers are nearly twice as likely as Millennials to feel negative toward AI ads (39% vs 20%), and that negativity is growing, up from a 21%-vs-15% split in 2024. (IAB, Jan 2026)
• 30% of Gen Z call AI-using brands “inauthentic” (vs 13% of Millennials), 26% “disconnected” (vs 8%) and 24% “unethical” (vs 8%). (IAB, Jan 2026)
• Gen Z excitement about AI fell 14 points in a year to 22%, while anger rose 9 points to 31%. (Gallup, 2025)
• 18% of US adults aged 18-29 would be more likely to buy from a brand using AI in ads, versus just 1% of over-65s. (CivicScience, 2025)
• Younger users are somewhat more forgiving of AI in the right format: two in five consumers aged 25-34 say they prefer generative-AI-powered creator content over traditional creator content. (Billion Dollar Boy, 2025)

4. Disclosure & transparency - and the paradox 

• 83% of consumers want disclosure when AI is being used, and 50% say an “AI-powered” label would actually increase their trust in the brand. (Emplifi, 2025)
• By 2026 that disclosure expectation reaches 91% of US and UK consumers. (Emplifi, Apr 2026)
• 28% of social media users worldwide say their single biggest brand turn-off is posting unlabelled AI content - the No.1 grievance. 50% of Gen Z have blocked, muted or unfollowed a brand or creator over “AI slop.” (Sprout Social, Q1 2026)
• 55% are more likely to trust brands that commit to human-generated content, rising to around two-thirds of Gen Z and Millennials. (Sprout Social, 2025)
• The paradox: clear disclosure is the third-strongest driver of attention to AI ads, and 73% of Gen Z/Millennials say knowing an ad was AI-made would increase or have no effect on purchase likelihood. (IAB, Jan 2026) Yet in controlled experiments, the same ad labelled “AI-generated” is rated less natural and less useful. (NIM, 2025)

5. The authenticity & trust penalty

• Awareness is shaky and trust is low: across the US, UK and Germany, only 44% know AI can create marketing content, just 25% think they can recognise it, and only 21% trust AI companies. (NIM, 2025)
• The “AI-authorship effect”: across seven preregistered experiments, believing an emotional marketing message was AI-written (rather than human) reduces positive word-of-mouth and loyalty - mediated by lower perceived authenticity and moral disgust. The effect shrinks for purely factual messages. (Journal of Business Research, 2025)
• Authenticity pays: 85% would pay more for brands they consider authentic, and the most-trusted signals are search results (66%) and user-generated content/reviews (63%), both ranking above brand-produced content. (Emplifi, Apr 2026)

6. The advertiser vs consumer perception gap

• The canonical gap: only 38% of consumers hold a positive view of AI, versus 77% of advertisers; and 65% of US adults are uncomfortable with using generative AI to create ads. (Basis Technologies, 2025)
• Marketers are scaling up regardless: 83% of ad executives say their company has deployed AI in the creative process, up from 60% in 2024, and AI is used most for social media ads (85%). (IAB, Jan 2026) 
• 73% of marketers and 78% of creators believe AI content performs better, while only 26% of consumers prefer it. That is a 44+ point professional-vs-consumer gap. (Billion Dollar Boy, 2025)

7. Creator economy impact

• 32% of US/UK consumers say AI is negatively disrupting the creator economy, up from 18% in 2023; the share viewing it positively (31%) is now at statistical parity. (Billion Dollar Boy, 2025)
• Supply and demand are diverging: 79% of marketers increased AI creator-content spend in the past year and 77% plan to divert more budget to it, even as consumer preference falls. (Billion Dollar Boy, 2025)
• 31% of consumers say AI in ads makes them less likely to choose a brand. (CivicScience, 2025)

8. When it goes wrong: brand backlash

• McDonald's Netherlands pulled its fully AI-generated Christmas ad within days of its December 2025 release after a wave of “AI slop” and “creepy” criticism, disabling comments before delisting it. (The Drum, Dec 2025)
• Coca-Cola ran a second consecutive AI-generated holiday campaign in 2025 despite 2024 backlash that branded the work “soulless” and “immediately forgettable.” (eMarketer, 2025)

9. Can people actually spot AI content?

• Only 50% of consumers correctly identified AI-generated copy in a blind test, and intriguingly, 56% preferred the AI version when they didn't know which was which. But 52% say they feel less engaged the moment they suspect copy is AI-generated. (Bynder, 2024)
• 56% of social users now see “AI slop” often or very often, and 66% have become more selective about what they engage with. (Sprout Social, Q1 2026)

10. UK vs US

• UK consumers were 10 points worse than US consumers at spotting AI copy (45% vs 55%), correlating with lower UK generative-AI usage at the time. (Bynder, 2024)
• UK trust in AI-generated content sits at 28%, below the US figure of 35%. (Emplifi, Apr 2026)
• The US and UK show notably lower concern about brands using AI than Canada or Australia, marking them as more “AI-ready” markets. (Attest, 2025)

From the field: what OOB sees in client work

The statistics above describe the market. The following reflects OOB's own experience

running AI-assisted content for clients; primary, practitioner observation rather than survey

data. We share it because it maps almost exactly onto the public research: AI wins in the utility-

and-amplification lane and loses in the authorship lane. Our operating principle: AI for amplification, not authorship; AI-executed, but creatively led by humans.

Where AI earns its place

In our work, generative AI performs strongly on topical, humorous, fast-turnaround content,

reactive memes and cultural moments where speed and timeliness are the entire

value. The economics are decisive: AI lets us produce something that looks genuinely premium

in hours, where commissioning designers, 3D artists and animators for a deliberately disposable

piece would be neither cost-effective nor sensible. Crucially, the idea is still human, a strong

creative concept, executed with the help of AI, not handed over to it.

A worked example

One calendar-moment activation which was AI-produced but creatively directed became, by a wide margin, the most-shared piece of content for that client to date, with a share uplift in the region of 1,700% against their benchmark.

Where we ring-fence it out

We deliberately keep AI away from genuine creative, cultural and identity work; no synthetic creators, no AI standing in for real creative craft. On the occasions clients have pushed AI into that territory against our advice, the content has consistently underperformed. That lived pattern is exactly what the wider data predicts: the 54%/13% creative-vs-shopping split and the 60%-to-26% collapse in preference for AI creator content are the macro version of what we watch happen at campaign level.

Indicative performance by content type

The takeaway we'd offer brands: AI has a real and valuable place, but it should amplify human creativity, not replace it. Ring-fence it out of anything where authenticity is the point.

A note on the data

Sample types vary and are not interchangeable. The Goldman Sachs figures come from a survey of the firm's summer interns - an elite, young population, not a representative consumer panel. IAB samples Gen Z and Millennials specifically; Emplifi samples frequent social-media users. Consumer positivity toward AI is variously reported at 38% (Basis, AI in ads), 45% (IAB, Gen Z/Millennial ads) and higher still when surveys ask about generative AI in general rather than in advertising, the figure depends heavily on what is asked and of whom. Much industry research is vendor-published and may carry mild self-interest; the academic and independent sources (NIM, the Journal of Business Research, Pew, Gallup, CivicScience) are useful counterweights for the most contested claims. Compiled by OOB (Our Own Brand). Statistics current as of 2025-2026; sources linked inline.