In the race for ad dollars, platforms seek to redefine themselves as AI companies
By Krystal Scanlon • June 25, 2025 •
Ivy Liu
Platforms are having an identity crisis. Or maybe, it’s a rebrand.
Whether they built their businesses on social graphs, short-form video or niche communities, the story from Cannes last week was the same: they’ve always been AI companies. That’s the new message being sold to marketers. The fact that AI rarely showed up in their pitch decks until recently seems beside the point. What matters now is being loud, and being first.
“It’s part of our DNA, which is why we’ve spent more than 10 years researching and investing tens of billions into our capex and building open source models,” said Meta’s vp of EMEA, Derya Matras.
Reddit made a similar case.
“We’ve always had machine learning and intelligence baked into everything we do,” said Reddit’s chief revenue officer Mike Romoff. “But with the launch of Community Intelligence powering one of our insights tools, empowering some of our ad products and helping optimize for outcomes, we’re really seeing AI run the gamut in our ads platform.” The insights tool Romoff is referring to is Reddit’s AI-powered social listening tool, Reddit Insights.
Every platform exec had their own version of this spiel on the Croisette. They’re trying to ride in the slipstream of a zeitgeist moment — one where AI isn’t just a tool, it’s the storyline. And in an industry where perception still sells, it’s no surprise that the platforms are repositioning long-standing machine-learning capabilities as proof they were early adopters all along.
And there’s some truth under the spin. AI has quietly powered core parts of the platform business for years — feed ranking, ad targeting, content moderation, translation and more. What’s changed is the marketing. AI is now the narrative, and platforms are scrambling to make sure their origin stories keep up.
“Innovation has always been a big part of what we offer, but I think what we’ve done in the last 18 months, from a marketer’s point of view, is combine innovation with performance,” said Snap’s chief business officer Ajit Mohan, referencing the early launch of Smart Campaign Solution — Snap’s answer to Meta’s Advantage+ — which makes it easier for advertisers to manage their campaigns in an automated fashion.
“There’s been a lot of investments in training models and ML to make sure we’re really able to be precise about improving ROI for advertisers,” Mohan added.
YouTube’s director of product management, YouTube ads surfaces and experiences, Melissa Nikolic, had a similar take, stating that she’s “really grateful” for the company’s AI capabilities because it “lets us solve these problems in a much faster and much easier way than before,” for advertisers and creators.
Even TikTok focused its energies on boasting about its latest Gen AI capabilities rather than entertaining any concerns about the latest in a long run of extensions of the TikTok ban.
“We’re integrating some of the Symphony features into Smart+, and we think that has big, big potential,” said TikTok’s head of creative product strategy and operations, Moritz Bartsch. “Smart+ has already automated a lot of things, made it very easy to run ads on TikTok, and that has a significant performance lift. So taking that a little further and bringing some of the creative elements into that is going to be the next step on that path.”
Still, not everyone is buying the repositioning wholesale. Some in the ad industry see the platforms’ AI push as an attempt to disintermediate agencies and vendors via automation. While some of that may happen as a side effect, it’s not the intent. Meta, Google and the rest have always treated holding companies and ad tech partners less as obstacles and more as extensions of their ecosystems.
What’s really happening is more pragmatic: these platforms are using AI to deepen their grip on their most important customer segment: small and medium-sized businesses. If AI makes it easier for those advertisers to spend more money, more often, then the platforms’ incentive to double down is clear.
https://digiday.com/?p=581616
