Are AI-generated images putting customers off?

Since generative AI entered the room, organizations around the world have experimented with hyper-personalized advertising, using the new technology to create thousands of versions of the same advert, tailored to specific cohorts of customers, segmented by persona. Other organizations have used generative AI as a way to reduce creative overheads, ditching the expensive shoots and visual design methods to create their brand assets with AI. And not because they are strapped for cash. Among the big names to go all out on generative AI are Coca-Cola, BMW, Moncler and McDonalds.

But, global research from Getty Images shows that although this method looks like the hot new way to engage customers or capture audience attention, it could actually be having the opposite effect. And not just because some AI-generated models end up with six fingers on each hand. 

In fact, Jacqueline Bourke, Getty Images’s senior director of creative for EMEA, says that despite the excitement around AI-generated imagery, the findings mean “brands should be thinking strategically about the use of this technology”.  

This article looks at some of the key findings from the research and answers the big questions about the use and future of AI-generated images in CX. 

Don't miss any news, updates or insider tips from CX Network by getting them delivered to your inbox. Sign up to our newsletter and join our community of experts. 

Learn More

Authentic imagery is essential to customer trust

In researching its VisualGPS report Building Trust with AI, Getty Images gathered perspectives from more than 30,000 adults in 25 countries between 2022 and 2024. It found 98 percent of consumers consider authentic imagery and video to be essential in earning their trust, while 78 percent agreed that, because of its origin, an image generated using AI cannot be considered “authentic”.

“That’s almost everyone, signalling that while AI brings incredible creative power, brands need to be thoughtful about the degree to which they use AI as a tool alongside human created content or a complete creative solution,” says Bourke.

“While consumers enjoy using AI to create images and art, they still view AI-generated works as less authentic and less valuable than human-made content, especially in advertising where they hold brands to a higher standard,” Bourke explains. 

Specifically, 66 percent of respondents said they prefer brands that use real, unedited photos in their ads, a finding Bourke says has been born of the consumer backlash different AI-generated ads have received. However, 74 percent agreed that an image made by AI can be just as realistic as a photograph taken with a camera, so to prove its findings the research also included image testing, during which AI imagery was more accepted by respondents when they thought it was real. When they found out it was AI generated, their acceptance dropped.

Getty’s isn’t the first study to identify these trends. In 2024, YouGov found that half of consumers say AI-generated ads are a turnoff.

Bourke says we’re seeing a “crisis of trust”. “Customers and consumers are constantly questioning what’s real, what’s edited, and what’s entirely synthetic,” she observes.  

Closing the trust gap around AI-generated images 

Given that 61 percent of those questioned by Getty Images expressed distrust towards all advertising – driven largely by concerns of AI-generated or manipulated imagery being used and misleading them – its report concluded there is a “critical trust gap in UK advertising for consumer-packaged goods”.

“Our research points towards one very important point; no one wants to feel misled,” says Bourke. 

“We see this in almost 90 percent of consumers wanting transparency where an image has been AI-modified or generated. We see it in the rejection of AI-generated people or product images. People want to trust that what they are being sold is what they will get. AI-generated visuals, by virtue of being synthetic, are not taken as a true representation of a product or service – so there’s an immediate disconnect.”

Despite 74 percent saying that AI-generated images can be just as realistic as authentic images, Bourke says people are more visually literate “than they have ever been” and as such are “hyper aware of images which look too polished or perfect and therefore [are] not an accurate representation of reality”. 

At the heart of it, she says, people want to see imagery that feels honest, a little imperfect, and recognizably real.

Many brands embraced this concept long before generative AI came on the scene. Beauty and fashion brands in particular have championed “real beauty” in the name of body positivity since the early 2000s. Nike’s inclusion of plus-sized models in its adverts dates back to 2016 the same year Sports Illustrated put a plus-size model on the cover of its swimwear issue. Summersalt’s “everybody is a beach body” campaign created waves in 2019, and other brands to embrace the movement have included H&M, Zara, and Barneys. 

Over this time other brands have hung their hats on “untouched” advertising imagery, among them Dove, Aéropostale's sister brand Aerie, Target, Bongo and CVS. 

Dove reported sales growth of US$1.5 in the first ten years of its "real beauty" campaign, which started in 2004 and, in 2014, Aerie saw a nine percent increase in sales quarter-on-quarter in response to its adoption of the trend. 

Rethinking how AI-generated images are used in CX

Untouched models, however, are not the same as AI-generated models. Since the 2022 arrival of ChatGPT, AI-generated images have frequently been seen by brands to be an effective and efficient way to generate huge volumes of visuals to support their campaigns. Yet the Getty Images research shows that consumers are more than twice as likely to prefer “real” images (38 percent) over AI-generated images (15 percent) in marketing communication materials. Does this mean brands using generative AI should rethink their creative strategies?

As Bourke said, brands should be thinking strategically about their use of this technology. “There are uses where consumers are more open to seeing AI generated imagery – for example, around conceptual visuals which are not people focused,” she explains.

“It is telling that the world’s biggest tech companies like OpenAI and Apple are grounding their recent campaigns in human craft – and being applauded for it – in comparison to beloved brands like McDonald's or Coca-Cola, both receiving backlash for ads generated entirely from generative AI,” she says. 

That said, there are markets where AI-generated images are not a turn off. Getty Images says consumers in markets such as Southeast Asia, Latin America, the UAE and Saudi Arabia are “much more receptive” to AI-generated imagery than in other global markets. 

Yet across industries, global consumers “prefer camera-based images” for marketing and advertising. AI-generated imagery was the least preferred type of imagery when Getty Images tested against “real” images, retouched “real images” and 3D/CGI images. 

“Healthcare marketing was the least accepted use case, with financial services the top use case – though again, low across all industries,” Bourke says.  

Using generative AI in customer journey

AI-generated content isn’t exclusive to the advertising element of the customer journey. It is used in post-sale service and support and is increasingly being used as a personalized shopping assistant in the research and discovery phase. Brands such as Amazon offer bot assistance to help customers navigate their vast catalogs, while OpenAI and Google have embedded generative AI shopping capabilities in their AI assistants. 

It’s clear AI has its place, but as always, to win and maintain customer trust brands must understand that customers are unwilling to accept AI as a like-for-like replacement for human contact or creativity. 

Bourke says: “Generative AI can be applied to great effect at many different points in the customer journey, driving efficiency and insights. To drive trust on the face of the brand, the technology needs to be used strategically and with thought given to how its audience will receive AI generated content. 

“People are already inundated with AI slop – low quality, mass-produced AI-generated content – so brands need to be careful that their own efforts don’t fall into this sea of sameness,” she adds. 

Quick links 

• ChatGPT's "Instant Checkout" lets shoppers buy inside chat
• The AI governance maturity model for CX leaders: Where does your organization stand in 2026?
• 7 ways companies are using generative AI in customer experience