When it comes to artificial intelligence (AI), it can be challenging to tell what’s real from what’s not. There’s a lot of hype, and brands that get caught up in it are playing with fire. How can AI brands – and brands that wish to use AI – protect themselves while taking advantage of the new opportunities the tech provides? In a fragmented, increasingly agentic shopping environment, trust becomes the differentiator – and IRL media are the channels that build that trust in public.
From Slopping to Shopping: How Consumers Encounter AI
The first thing to understand is that AI-generated content is everywhere.
More than half of the new articles on the web are now AI-generated. On digital platforms, the problem is even worse. Spotify had to delete 75 million AI-generated songs last year, and after YouTube was flooded with algorithm-gaming AI slop, its CEO stated that managing the problem is its top priority for 2026. TikTok, where there are more than 1.3 billion AI-generated videos, has already given users some control over how much of it shows up in their feed.
It’s no wonder that Merriam-Webster named “slop” its 2025 word of the year. But there is one scenario where consumers are actually surprisingly receptive to AI-generated content: shopping.
Nearly four in ten shoppers use AI, and nearly half of those who do, do so every time. Comparison shopping and product-specific questions are the most popular shopping tasks assisted by AI.
Perhaps one reason for AI’s acceptance here is that implementations like review summaries are generated passively. But shoppers’ embrace of AI is not universal across the board. Personalized recommendations are not catching on as quickly, and trust is a big reason why. Just 36% of consumers say they trust AI-generated shopping recommendations. But when those decisions are explained, that number rises to 66%, according to technology consultant Capgemini.
The Productivity Paradox and the Trust Gap
On the business-to-business side, the picture is also mixed. While OpenAI claims that the average ChatGPT Enterprise user saves 40 to 60 minutes daily, there is reason to be skeptical of the claim.
A recent Wall Street Journal survey revealed that 40% of workers who use AI save no time at all, while PwC’s annual CEO survey revealed that 56% of companies are seeing no cost or revenue benefits. And just in the last few weeks, the National Bureau of Economic Research discovered that nearly 90% of companies implementing AI have seen no impact on employment or productivity, in part due to the ease of creating “workslop” time-sinks that recall economist Robert Solow’s productivity paradox, first observed in 1987 alongside the advent of personal computers.
The distance between the promises of AI boosters and the reality of what the technology can do has opened up a trust gap. Only 3% of us trust AI a lot. Nearly 70% say the information it generates isn’t trustworthy, and 71% are concerned about the lack of clarity in how AI systems collect and use personal data.
The externalities of AI are hard to ignore, from higher power bills to PC hardware shortages. And its deeply-embedded penchant for hallucination is another reason why consumers are wary of using the tech. Data centers’ impact on communities is another big reason. In fact, local opposition has derailed or delayed $98 billion worth of data center projects nationwide in Q2 of last year.
Besides those negative consequences, there’s also the impact AI makes on the minds of those who use it. A study by the MIT Media Lab revealed that “LLM users consistently underperformed at neural, linguistic and behavioral levels.” The study describes “an accumulation of cognitive debt” and “low effort, mostly copy-paste” output.
Even worse, reliance on AI has the potential to make us worse at being real humans. The majority say that people’s ability to do things on their own will suffer from using AI, reports MIT Media Lab. Overwhelmingly, we fear that it’ll make us worse at things like thinking creatively, forming meaningful relationships, making difficult decisions, and solving problems, according to Pew Research.
And then, alarmingly, there’s the psychosis. More than a half-million ChatGPT users show signs of psychosis or mania each week, reports British medical journal The BMJ, and AI companion products have been implicated for their role in 11 suicide deaths, according to Public Citizen.
The Way Forward for AI Is Earning Trust
We’re at the point where we need to figure out how and where we want AI to fit into our lives
At the end of the day, our relationship with AI comes down to trust. It’s an uphill battle for AI brands, but not an insurmountable one. There is a path forward.
To earn trust, brands selling AI can emphasize three important pillars of trust: transparency, explainability and human involvement. Some 62% of respondents say they’d have more trust in brands that are transparent about their use of AI. Some 83% would trust AI more if it could explain its reasoning or decisions. And 78% would trust it more with humans in the development loop, reports RWS in its report, “Unlocked 2025: Riding the Shockwave.”
There’s no denying that AI is powerful, or that it’s rapidly changing the world around us. But whether or not “AI is here to stay?” That depends on closing the trust gap.
One way that AI startups have been addressing the trust gap has been to establish a tangible real-world presence with IRL formats like billboards and subway posters. That physical presence builds more than name recognition – it tells your audience that you’re credible. It communicates that you’re serious, dependable and worthy of their trust.
After all, in a world where everything seems fake, nothing wins like getting real.
For more on this, check out Outfront’s 2026 Advertising Trends Report. The report covers consumer trends, the advertising media landscape, and AI, and it shows how IRL media connect with consumers in ways online formats cannot match.
Jay Fenster is senior manager, content marketing, at Outfront, one of the U.S.’s largest and most trusted out-of-home media companies , helping brands connect with audiences in the moments and environments that matter most.
