Generative AI in Video Ads: Key Findings
86% of U.S. buyers use or plan to use generative AI in video ads, showing why brands must adopt AI workflows to stay competitive on speed and cost.
AI-powered YouTube campaigns deliver 17% higher ROAS than manual ones, proving that automation outperforms when paired with strong creative strategy.
Mondelez is slashing production costs by up to 50% using AI with Publicis and Accenture, showing how brands can reinvest savings into sharper storytelling and bigger ideas.
Generative AI has made video faster and cheaper to produce. That part is obvious.
What follows is less straightforward; feeds are filling up with automated content that looks complete but feels interchangeable.
Now, creative judgment carries more weight than output.
Taste, intent, and restraint shape how the work lands.
In an interview with DesignRush, Chris Marcus, Chief Creative Officer at Colormatics, explains how brands can use AI while protecting creative judgment and human presence.
Who Is Chris Marcus?
Chris runs the creative process at Colormatics, directing national commercials for brands like PrizePicks, Squatty Potty, Fiverr, and Sleeper. He has worked with Stephen Curry, Shaquille O’Neal, and Trevor Lawrence, helping the agency compete with major players in advertising.
Editor’s Note: This is a sponsored article created in partnership with Colormatics.
Generative AI is no longer just an idea.
Nearly nine in 10 U.S. buyers are already using or planning to use AI for video ads, according to the IAB 2025 Digital Video Ad Spend & Strategy report.
And small and mid-sized brands are adopting it faster than the largest enterprises.
What’s driving AI adoption? Often, it’s cost savings.
In October 2025, Reuters reported that Mondelez, maker of Oreo and Chips Ahoy, began using a generative AI tool built with Publicis and Accenture to cut video production costs by up to 50%.
The same Reuters reporting states that a Mondelez executive expects the tool will be able to generate short TV ads ready to air as early as the 2026 holiday season and potentially for the 2027 Super Bowl.
Additionally, market data is nudging marketers further into automation.
Google found that AI-powered YouTube campaigns deliver 17% higher return on ad spend than manual campaigns, based on a 2025 Google and Nielsen analysis.
Which means brands are saving time and cutting costs. That part is easy.
But speed and efficiency don’t tell the full story.
The bigger risk is losing the human judgment that makes content memorable.
https://www.instagram.com/reel/DRnWIVKE4mQ/embed/captioned/?cr=1&v=14&wp=810&rd=https%3A%2F%2Fnews.designrush.com&rp=%2F86-use-generative-ai-video-ads-creativity-still-wins#%7B%22ci%22%3A0%2C%22os%22%3A3618.100000023842%2C%22ls%22%3A1108.6999999284744%2C%22le%22%3A1127.6000000238419%7D
According to Marcus, the most common mistake brands make with AI video has nothing to do with technology.
“They confuse efficiency with effectiveness. AI is being used to make more content, faster, cheaper, and safer. That is the wrong goal,” he says.
“The result is a flood of visually competent, emotionally vacant work that all feels the same. Brands think they are scaling creativity when they are really scaling mediocrity.”
This played out in real campaigns.
In 2024 and 2025, Coca-Cola faced backlash for a fully AI-generated version of its iconic “Holidays Are Coming” campaign.
Critics said it lost authenticity and emotional impact, even though it was faster and cheaper to produce.
AI can speed production, but it can’t replace human taste and judgment.
Speed is the easy win. The harder question is how to put that extra time and budget to work.
What Brands Should Focus on Instead
Marcus believes the smartest brands are rethinking where AI fits in their creative process.
“Use AI to compress the boring parts,” he says, pointing to testing, versioning, animatics, and exploration,” Marcus says.
“Then reinvest that saved time and money into sharper ideas, better casting, and braver creative decisions.”
With generative content flooding feeds, volume no longer builds relevance.
As Marcus puts it:
“Culture rewards taste, not output volume.”
AI should amplify human judgment, not replace it.
The numbers back it up.
A 2025 Forrester study found that more than 60% of U.S. agency decision-makers are already using generative AI, with adoption rising to 78% among large agencies.
Those who are using it well are restructuring workflows to support creativity, not cut it out.
3 Practical Steps for Sharper Visual Work in 2026
Looking ahead, Marcus points to three actions brand teams should take if they want their work to hold attention.
1. Lock a Visual Point of View
Brands need to know what they look like even without a logo on screen. That includes camera choices, pacing, color, and performance style.
“If it cannot be explained on one page, it is not real,” Marcus says.
In other words, if your identity isn’t clear, consistent, and creatively useful, it won’t show up with any power in the work.
2. Cast for Energy, Not Perfection
Real faces, imperfect timing, and human presence consistently outperform polished, synthetic visuals in crowded feeds.
3. Design for Attention Decay
Assume you have one second. Open strong enough to earn the next two.
These choices matter more as spending climbs.
U.S. digital video ad spend grew 18% year over year to $64 billion in 2024 and is projected to reach $72 billion in 2025, according to the IAB.
But don’t be fooled.
Even with rising budgets, the goal isn’t more content; it should always be better content.
Despite the speed of change, Marcus says many brands are still holding back where it matters most: tone.
“Everyone is afraid of being polarizing, so everything becomes pleasant.”
The result is work that fades into the feed instead of cutting through it. Humor gets diluted. Emotion gets softened. Risk quietly gets handed off to algorithms.
Ironically, he argues, “The safest thing now is being bold, because sameness is the riskiest position in the feed.”
Examples like Mondelez show how far efficiency can go.
AI-driven production has helped large brands cut costs and increase output at scale.
Balancing Efficiency With Emotional Depth
The risk, Marcus says, is assuming those gains should apply to every part of the process.
“Separate what must be efficient from what must be human,” he advises.
Production mechanics can be optimized. Story, emotion, casting, and taste cannot.
When AI starts shaping the emotional core of the work, brands may save money in the short term while wearing down long-term brand value.
“Emotional storytelling is not expensive. It is rare. There is a difference,” Marcus says.
Marcus echoed this theme in a recent panel discussion, where he spoke about how AI can support storytelling but cannot replace the creative decisions that give work meaning:
As short-form, AI-generated ads take over social feeds, Marcus says brands need to rethink how different types of content work together.
“Think in layers, not volume,” he says.
- Fast, lightweight content should help test ideas and maintain presence.
- Fewer, higher-impact pieces should do the heavier lifting by anchoring the brand’s identity.
Short-form content wins clicks. Longer-form storytelling builds belief.
“Brands that only do one end up renting attention instead of owning it,” Marcus says.
And so, smart brands use short-form to stay visible and long-form to stay remembered.
Looking at Colormatics’ recent work, Marcus sees a consistent pattern.
The campaigns that perform best are not the most optimized. They are the most intentional.
“When the idea is clear, the tone is confident, and the execution feels human, it cuts through regardless of platform,” he points out.
AI can help teams move faster and test more, but the real gains come from taste, conviction, and restraint.
That level of work also demands change inside organizations.
Marcus points to smaller teams, senior decision-makers closer to the work, and clearer creative ownership.
As AI lowers the barrier to production, leadership and judgment become the real differentiators.
