How AI could shape content and ads in 2025

Tech giants and startups alike have spent the past year building new generative AI tools for users and advertisers.

From AI images for programmatic ads to a growing number of AI-generated TV commercials, brands are starting to explore new ways of thinking about creative across various platforms. The final weeks of 2024 had big news with expanded access and improved outputs of generative models like OpenAI’s Sora, Amazon’s Nova and Google’s Veo.

Despite the technical feats, AI-generated content has gained both eager devotees and harsh critics. Depending who you ask, the category’s a powerful new form of creativity, underwhelming “AI slop,” or an IP-stealing job-killer. However, the question is, which of these viewpoints will be the one to stick.

In the coming year, those tools and others from companies like TikTok, Meta and Amazon could see more adoption with users, influencers and advertisers. Better results and lower costs could lead to a surge in AI-generated content for users and businesses alike. However, it’s unclear how regulations and litigation related to copyright issues could impact innovation, adoption and creation in 2025.

One startup to gain praise is Pika, which recently released its AI video model Pika 2. Lindsay Brillson, Pika’s head of brand and content, said Pika’s seen more brands using it for organic social content and sometimes for their ads. “It’s fun and it gets people’s attention and it feels native to the space,” said Brillson. 

“There’s a sea change to be able to really control the building blocks of the scene and not just leave it up to chance,” she said. “Now we know that people want to be able to control the characters, the background, the wardrobe, the objects, the props, the products.”

Platforms like Pinterest, Meta and Snap have spent the past year building out new ways for advertisers to create AI-generated ads while also rolling out new features for users. Content creation tools for users and businesses are bleeding together in ways that could help drive more adoption, said Kevan Yalowitz, global industry lead for software & platforms at Accenture. He’s also excited about the potential benefits of more personalized and dynamic ads on social platforms and elsewhere.

“For the first time ever, they can actually create an ad that is directly tailored to that individual user, but also the individual query,” Yalowitz said. “That went from idea to reality on most platforms this year.”

SMBs will continue adopting AI content and ad tools from social platforms, said Jon Morgenstern, evp and head of investment at VaynerMedia. He mentioned TikTok’s Symphony Creative Studio that lets people paste a URL into the tool to create AI-generated content. Larger brands are hesitant to adopt certain AI tools due to concerns about data privacy, copyright infringement and intellectual property, which could give smaller brands an edge. 

“The long-tail is using these [tools] immediately,” Morgenstern said. “It’s kind of leveling the playing field almost disproportionately, where challenger smaller brands can fly under the radar and not get sued for doing things versus the big guys.”

Generative AI’s impact on creative assets has yet to fully be felt, said Barney Worfolk-Smith, chief growth officer at DAIVID, which uses AI measure how ads gain attention and affect emotions. He sees the potential of AI to help with predictive analytics and synthetic audiences, but thinks AI’s benefits for personalization and scale are overblown and could detract from quality outputs. However, that could change as outputs improve and costs decline.

“We compared the emotions of Coca-Cola’s human-made versus AI-made,” Worfolk-Smith said. “Warmth, which is a standard emotion evoked within Christmas ads, was about a third lower [in the AI ad]. We essentially attributed that to the uncanny valley.

While platforms are giving new AI tools for creators to use, some question how content creators will be compensated if their content’s used to train AI models that others turn into content. Gartner analyst Andrew Frank mentioned efforts underway by companies like Bria, Getty Images and Adobe that seek to pay contributors based on their impact on AI-generated outputs. 

“The overall mood is that we’re at a high point of anxiety, not only because of politics, but because things have been changing so quickly,” said Frank. “I think there’s still a lot of future shock from a lot of the developments of this year, including AI developments and developments in other technologies … People are sensing that things are moving very quickly and they’re sort of out of control.”

Whether control will be achieved in 2025 remains to be seen. But there’s no doubt the envelope tech providers, adopters and humans alike will continue finding new ways to develop, deploy and test new types of AI-powered creative.

https://digiday.com/?p=564415

More in Media

AI Briefing: Writer’s CTO on how to make AI models think more creatively

Waseem Al Sheikh, Writer’s co-founder and chief technology officer, talks about how to help LLMs think more like humans — and make fewer mistakes.

Generative AI grows up: Digiday’s 2024 timeline of transformation

Alongside innovation, partnerships and adoption, Year 2 of the generative AI race also includes more scrutiny over transparency, copyright, and ethics.

Digiday+ Research roundup: Publishers’ revenue tactics and TikTok were 2024’s biggest topics

We rounded up the biggest trends of the year, based on the data that resonated the most with our readers.