The year AI reshaped marketing and media

If 2024 marked AI’s arrival in mainstream marketing, 2025 was the year it rewired the system. Global brands recast creative campaigns, agencies reorganized around automation and publishers grappled with evaporating referral traffic as LLM-driven search took center stage. AI became the gravitational force exerting pressure on every corner of the industry.

Three themes defined the year: AI became the norm for business, moving from experimental to essential, with major brands investing heavily in AI stacks; agencies restructured to integrate automation at scale; and LLMs brought search to a whole new level, putting even more pressure on publishers.

For brands, AI moved from experiment to expectation

By 2025, AI had moved out of pilot mode for major marketers — it became a frontline tool shaping how brands show up, from storytelling to distribution to customer interaction. The most telling AI signs came from the world’s largest marketers, who made it clear that generative technology is no longer optional, experimental or hidden behind the scenes.

Just this month, Disney announced a $1 billion equity investment in OpenAI, enabling users of OpenAI’s Sora and ChatGPT to create short videos using iconic characters like Mickey Mouse and Cinderella. Meanwhile, OpenAI’s chatbot app program attracted brands like Spotify, Expedia, Zillow, Canva and Adobe.

AI’s tech is also being used to create ads. Coca-Cola set the tone early, returning with another major AI-driven holiday campaign — this time produced with generative tools playing a central role. Despite mixed reactions from viewers, the company made its intention unmistakable: AI is now part of Coca-Cola’s creative operating system. Its co-creation platform invited fans to generate branded imagery, signaling a shift toward scalable, user-assisted brand expression. This holiday season, the trend continues with “Refresh Your Holidays” from WPP Open X and VML.

Mondelēz International, owner of Oreos and Cadbury, made perhaps the most consequential structural move. Investing more than $40 million in a proprietary generative-AI production engine, the company committed to using AI to generate everything from social assets to full broadcast-ready commercials. Early estimates suggest a potential 30–50% reduction in creative production costs — a number that caught the attention of other global advertisers.

The company’s plan to use the system for tentpole moments, even Super Bowl-level placements, demonstrates how deeply AI may permeate brand production in the coming years.

Still, the year saw plenty of growing pains show up. Google’s Gemini Super Bowl ad errors, Valentino’s social AI-generated ad and Duolingo’s over-eager jumpstart into AI highlighted the risks. Overall, there’s more optimism than fear, and platforms like Meta and Canva are spurring AI adoption with more tools for marketers and creators. The new year will likely see more brands jumping into the fray, such as the vodka brand Svedka which just announced an upcoming AI-assisted Super Bowl spot.

Agencies: Consolidation, automation and the new AI stack

If brands moved boldly, agencies reorganized even faster.

Holding companies moved quickly to re-architect themselves around AI, rolling out proprietary platforms like WPP Open Pro, Publicis’ CoreAI and Omnicom’s Omni Assist, while appointing dedicated AI leadership and reorganizing creative, media and production teams to scale automation.

But the challenge was not merely technological — it was existential. As brands invested in their own generative tools and co-creation platforms, agencies had to redefine their value beyond execution, emphasizing marketing strategy as their workflows become more automated.

Meanwhile, a new class of AI-native specialty shops emerged, pitching “co-pilot” production models, hyper-personalized creative variants and AI-accelerated concepting. These challengers put pressure on the traditional agency model.

Arguably the largest instance of AI toppling traditional agencies was the merger of Omnicom and Interpublic Group towards the end of this year.

In an op-ed for The Current, Sam Huston, SVP, creative and media at ad agency DEPT, described the merger in this AI era: “The merged company is architecting for automation across everything that can be templatized: optimization, forecasting, reporting, content production, QA, trafficking and a growing share of lower-level creative execution. Work that once required teams of 10 could require teams of three, assisted by machine-learning models that don’t need coffee breaks.”

LLMs and search: The traffic reversal hits publishers

While brands and agencies debated efficiency and creativity, publishers faced a far more existential AI development.

LLM-driven search and answer experiences became default across major platforms, sharply reducing referral traffic for publishers. Consumers increasingly relied on AI-generated summaries or synthesized answers, rather than clicking through to source sites. At the same time, newsrooms began integrating LLMs into their workflows, further blurring the line between original journalism and AI-assisted production.

Publishers saw that it was time to recalculate how to build ad revenue, with some exploring partnerships with LLM platforms.

“Publishers should treat AI not as a threat, but as a new monetization layer — one that complements existing revenue streams like subscriptions, advertising and syndication,” wrote Josh Peters, Bombora’s VP of global data partnerships and strategy, in a 125-page guide he created on setting an AI monetization foundation.

What this means for 2026 and beyond

Across industries, a cultural shift took place: AI became a material, not a gimmick.

The tension between efficiency and artistry sharpened. Viewers criticized some AI-generated campaigns as “soulless,” while brands praised the technology’s speed and scale. Yet the debate itself confirmed AI’s new status.

Taken together, the year’s movements point to a restructuring of marketing and media. And the ripple effects will define the competitive landscape for years to come. As Terence Kawaja, founder and CEO of LUMA Partners, put it — expect more consolidation ahead.

“It’s going to be like a sword slicing through the LUMAscape because AI accelerates the move to outcomes,” Kawaja told The Current.