Generative artificial intelligence has made a dramatic entrance into the world of marketing and advertising. And it’s continuing to raise concerns. We take a closer look...
By Bertrand Beauté
Image credit: MANGO
In the height of summer, while all eyes were on the Olympics, the Spanish ready-to-wear fashion giant Mango found a way to make headlines with a new kind of advert. The image shows a dark-haired young woman with a souk in the background to promote Mango’s new Sunset Dream collection for teens. At first glance, there’s nothing out of the ordinary about the model. She’s the embodiment of the ideals of women’s fashion. Except that neither the woman nor the souk exists. In a press release published on 10 July, Mango celebrated the launch of the first "campaign generated entirely by generative AI". In actual fact, Mango isn’t the first company to use AI to make an advert. In 2022, for example, advertising agency Ogilvy Paris used AI to extend Johannes Vermeer’s painting "The Milkmaid" in an ad for Nestlé’s La Laitière brand.
"Generative AI has a real impact on content creation. We’re seeing more and more adverts made entirely using AI," says Alexandre Zilliox, portfolio manager at Thematics Asset Management. Figures show that AI is becoming more prevalent in the sector. For example, in 2021, just 29.7% of marketing experts surveyed used AI in their work, but by 2023, this figure had already risen to 40.6%, according to "How Artificial Intelligence Is Revolutionising Content Marketing", a study carried out for the fourth time in 2023 by SRH Berlin University of Applied Sciences on behalf of Statista.
Mango’s experiment was intended to encourage young people to think of the company as a passionate adopter of new technology, but AI offers many other benefits. Advertisers can make substantial savings because they no longer need to pay models, photographers and graphic designers or organise photoshoots on the other side of the world. Let’s look at an example. If a car manufacturer wants to promote its newest model, very soon, it won’t need to send an entire crew to film it driving in the middle of the desert. "AI can also be used to automatically generate several visuals for a single campaign, or create a video version of a print advert," explains Jean Meneveau, director at Colombus Consulting’s Swiss office.
The big web players have understood AI’s huge potential in the advertising sphere. In October 2023, Amazon launched a tool that allows sellers to create advertising visuals using artificial intelligence. The Seattle-based company had previously used AI to generate product descriptions and provide customer service. Meta and Google are following the same trend. In May 2023, Meta – the parent company of Facebook, Instagram and WhatsApp – revealed its AI Sandbox, which allows advertisers to automatically generate ad images and text with AI. Google made the leap a few days later with its own tool for automatically generated adverts.
Ben James, from Baillie Gifford
"Historically, Google wasn’t focused on creativity. That was the preserve of advertising agencies," says Zilliox. "But with AI, that could well change." Could this push major agencies such as Publicis, Omnicon, WPP and the like into bankruptcy by robbing them of some of their business? It’s a question that divides the experts we spoke to. "At the moment, agencies are still key players, but perhaps advertisers will use AI tools directly one day," says Humberto Nardiello of DPAM.
He continues: "The risk for traditional ad agencies is that ultimately, AI could allow advertisers to bring part of their work back in-house. That’s one possible scenario." Ben James from Baillie Gifford agrees: "AI is a gamechanger in advertising. Big tech companies are ideally placed to benefit from it while, on the other hand, agencies are at risk of losing their work."
Ludovic Labal from Banque Eric Sturdza is more moderate. "I don’t think AI’s going to kill creativity. Look back 30 years when Excel arrived on the scene. At the time, everyone said that accountants were going to disappear. But in the end, there were actually even more accountants than before. Why? Because when the price of something collapses, people use it much more. We’re going to see more advertising because AI’s going to make it cheaper. And that’s actually pretty good news for agencies."
HYPER-PERSONALISATION
Although AI threatens to disrupt creativity in advertising, it will have a huge impact on other elements in the sector. According to "How Artificial Intelligence Is Revolutionising Content Marketing", AI is affecting six marketing aspects: content creation, personalisation, customer action forecasts, sentiment analysis, better use of data and search engine optimisation.
"The real revolution lies in using AI to create hyper-personalisation," says Jean Meneveau. Rather than showing the same ad to everyone, the idea is that every consumer sees a version that’s personalised to them. "Extreme personalisation is one of the things AI is promising. It’s a dream for advertisers and a nightmare for consumers," says Labal.
Imagine if tomorrow, instead of seeing a model wearing the outfit you’re thinking of buying, it was your own face and body in the ad. "Personalisation is one of the aims of AI, but it’s also a risk," James warns. "If an ad is tailored too closely to an individual, what impact will that have on a consumer’s free will?" We’re not there yet, but there have been several attempts
to move in that direction. For example, in 2023, the American cruise line Virgin Voyages launched an ad campaign in which web users could send their friends and family a video of Jennifer Lopez inviting them to book a cruise. And they could even personalise JLo’s message with AI – a kind of deepfake with a Virgin twist. "AI is particularly useful for creating multiple iterations for a single campaign," adds Alan Mudie from Woodman Asset Management.
In 2019, the American bank JPMorgan Chase signed a partnership with Persado to use AI to design more effective, more personalised marketing messages aimed at its customers based on their banking activity. And America’s biggest bank isn’t stopping there. In early 2024, it announced it would become an ad agency. The bank promises brands the ability to target its millions of customers with extreme precision through its credit card spending data. As a bank, it knows exactly where, when and at what store its customers usually shop. It’s a treasure trove for advertisers, meaning that they can target the right ad at the right time via the best channel.
CUSTOMER DATA: THE HOLY GRAIL
It’s an example that shows just how important it is to have a large customer base. "Data is fundamental in the world of advertising, and big companies have a clear advantage because you need a lot of data to be relevant," says Labal. And the result is that, in addition to advertising giants like Google and Meta, other companies are increasingly processing their data using AI to benefit advertisers.
Ludovic Labal, portfolio manager at Banque Eric Sturdza
Alongside JPMorgan, there are e-commerce sites (Amazon, Walmart, Alibaba), food delivery companies (Uber, Deliveroo) and streaming sites (Netflix, Spotify). Ad agencies haven’t stood idly by in the face of this phenomenon, but have increased their data-related acquisitions. For example, in 2019, Publicis spent $4.4 billion on the American firm Epsilon, the second-largest digital marketing and customer relationship management (CRM) company in the world. "Agencies are going to keep on buying digital companies," Labal predicts. And for now, it’s working. Far from experiencing a crisis, Publicis is posting record results. Its share price is also soaring: up 34% year-on-year, with the leading French ad agency breaking the symbolic €100 threshold for the first time in its history at the end of March before dropping to around €95 in mid-August.
This robust performance comes as no surprise to Jean Meneveau: "The big agencies have always excelled in terms of innovation. In recent years, they’ve also bought a lot of data players. It’s 2024, and they definitely know how to process data." But although they have the data, agencies don’t have big tech’s raw power when it comes to AI. "I’m not convinced that ad agencies have the resources to compete with big platforms in the long term when it comes to generative AI. They’ll have to use tools from big tech, and maybe they won’t be able to charge as much for their services as before," says Alexandre Zilliox of Thematics Asset Management. Especially because campaign planning is another aspect that’s increasingly assisted by AI, which determines where and when to display an advert and boosts campaigns’ click rates, engagement rates and conversion rates. The fact remains that the rise of AI in advertising isn’t without its ethical issues. After launching its campaign, Mango came under criticism from many, who condemned the brand for using AI to reinforce clichés and set increasingly unrealistic beauty standards. What’s more, "AIs can present biases based on the data they use, which can lead to certain demographic groups getting preferential treatment while others are discriminated against, and this poses a problem in terms of marketing and advertising," says Ben James from Baillie Gifford. But in an attempt to avoid this pitfall, Google went too far the other way. In early 2024, its AI tool Gemini was criticised for generating images of Black and Asian Nazi soldiers.
And automatic content generation isn’t the only problem. "AI works to optimise campaigns," James points out. "In the health-care sector, would an AI-led campaign target the people who need a particular medication the most, or those who can afford to pay for it?"