Gen Z Is Punishing Brands for AI-Generated Marketing. The Problem Is Not the AI

A study published on 13 August found that 72% of Gen Z consumers have taken direct action against a brand after realising its marketing was made with AI, from unfollowing and unsubscribing to complaining publicly and stopping buying altogether. The research, from Rival Technologies, surveyed 901 Gen Z participants across the US and Canada in July 2026: half had unfollowed a brand, 49% had complained to friends, family or online, 48% had unsubscribed and 43% had stopped purchasing entirely. It landed two days after Google confirmed that its Gemini app had passed a billion monthly users. Adoption and trust are moving in opposite directions, and the distance between them is the real story.

What did the research actually find, underneath the headline number?

Not that people dislike AI. That they dislike discovering it.

The same study found 74% react negatively on realising a brand's marketing was made with AI, with half describing that reaction as very negative, and only 8% reacting positively. The trigger is the realisation, not the technology. Nobody is auditing a brand's tooling. They notice that something feels synthetic, and then they feel misled.

That distinction matters commercially, because it is the difference between a problem you solve by not using AI and a problem you solve by being straight about how you use it.

Why is the gap between what advertisers believe and what consumers feel the more important number?

Because it is widening, and somebody is now measuring it. IAB research with Sonata Insights found that 82% of advertising executives believe Gen Z and millennial consumers feel positively about AI-generated ads, while only 45% of those consumers actually do. That perception gap has grown from 32 points in 2024 to 37 points in 2026. On the same data, Gen Z is nearly twice as likely as millennials to view AI-generated advertising negatively, 39% against 20%.

A 37-point gap between what an industry believes about its audience and what that audience reports is not a technology problem. It is a listening problem, and it has the shape of every reputational failure I have worked on: the organisation is confident, the audience is not, and nobody inside the organisation is measuring the difference until it turns up in the sales figures.

Is the problem AI itself, or the absence of disclosure?

The evidence points fairly clearly at disclosure. eMarketer's analysis found that 44.1% of adults said they liked ads more after discovering AI had been used, including 49.5% of Gen Z, and that 73% of Gen Z and millennial consumers said knowing an ad had been made with AI would either increase their likelihood of purchase or make no difference to it.

Set that against the 72% who have punished a brand and the picture resolves. Disclosed AI is largely tolerated and occasionally rewarded. Undisclosed AI, found out by the audience, is punished. The IAB has now published an AI transparency and disclosure framework for precisely this reason, which is a reasonable signal that the industry has accepted disclosure, rather than usage, as the live variable.

For communications teams the practical reading is uncomfortable and simple. The risk does not sit in the workflow. It sits in the gap between what you used and what you said you used.

What does this mean for earned media specifically?

Earned media is the one channel this backlash cannot reach, and that is worth stating plainly rather than defensively.

A journalist's write-up of your company cannot be generated. It is the product of somebody independent deciding you were worth putting their name to. That is exactly the property Gen Z is reaching for when it punishes synthetic marketing: evidence that a human being with something to lose has vouched for you.

It is also why the AI-era launches that have worked best are the ones that put the human outcome in front of the technology. The MyHeritage Deep Nostalgia launch worked because it was framed as an emotional experience, seeing a relative move again, rather than as a machine-learning feature. The coverage that followed was earned, across broadcast and national press, not generated, and that is why it carried.

What should a communications team change this quarter?

• Write down where AI is actually used across your marketing and communications output, honestly, including drafting and image work. Most teams cannot currently answer this, which makes any public statement about it a risk.

• Decide your disclosure line before you need it. Not a policy document, a single sentence you would be comfortable seeing quoted in a story about your brand.

• Keep AI out of anything trading on authenticity: founder voice, customer stories, testimonials, condolences and apologies. The efficiency saved there is small and the exposure is total.

• Rebalance the mix towards what cannot be synthesised. Earned coverage, named spokespeople, real customers on the record.

• Prepare for the question. 'Was this made with AI' is now a live media question, and a hesitant answer is the story.

Where does that leave AI inside the PR workflow?

Firmly in place and genuinely useful, provided it stays behind the work rather than in front of it. AI is good at research, media monitoring, finding patterns across coverage, first-draft structure and speed. It is not good at judgement, at relationships, or at the specific decision about what a story will mean to a particular journalist on a particular day.

The line I hold as a Senior Freelance PR Consultant is that AI can shape how I work and never what a client says in public without a person answerable for it. On the evidence above, that is not caution. It is the position the audience has already taken, and the brands still arguing with it are arguing with 72% of a generation.

If your marketing has moved faster on AI than your communications position has, talk to Fireflies Management.