How AI in Advertising is Turning Big Brands into Creative Experiment Labs

Everyone talks about AI “arriving” in advertising like it’s a future threat. The truth is, it has already stormed in, upending everything we thought we knew about creativity and campaigns.

From Coca-Cola’s AI-powered holiday ads to Amazon’s generative video tools and Coign’s fully AI-created TV commercial, one question can’t be ignored: if AI can imagine, optimize, and produce ads, what happens to the human spark that makes creativity unforgettable?

In this blog, we explore how AI is reshaping advertising, where it excels, and why human insight remains the ultimate differentiator.

Coca-Cola’s Generative AI Experiment: Reimagining a Holiday Classic

Coca-Cola​‍​‌‍​‍‌​‍​‌‍​‍‌ has been the talk of the town with its holiday campaign for 2025, which reimagined its iconic trucks and festive vibes with artificial intelligence. The brand invited AI to remix its iconic imagery, creating visuals that felt part human, part machine. 

It used 70,000 generated video clips refined by a handful of AI specialists. In the ad, sparkling Coke bottles floated in surreal landscapes. It looked magical. People shared it, and the press wrote about it. And yes, it generated a lot of buzz.

Coca-Cola’s holiday campaign showcased the advantages of AI video production: speed, scale, and cost-efficiency. The brand, armed with creative AI tools, was able to do in weeks what usually took months, keeping up with cultural trends in real time.

In such a production environment, thousands of visual variations are made possible, giving teams freedom to test, refine and adapt in different markets as needed. With reduced dependence on physical shoots and logistics, the campaign achieved remarkable cost efficiency: AI was turned into both a creative and an operational advantage.

But the very tools that enabled this speed and scale also had their limits, with critics highlighting some uncanny visuals. Eagle-eyed viewers quickly spotted odd glitches in Coca-Cola’s AI-generated video ad. The iconic red trucks seemed to change shape as they moved through the snowy village, even gaining or losing wheels between scenes.

These inconsistencies exposed a key weakness of current generative video models: they struggle to maintain visual continuity across shots. Since many systems generate frames independently without remembering prior ones, the result is what experts call “temporal drift”,  a telltale sign that AI still can’t fully match human precision in storytelling.

Amazon: Efficiency in AI Video Ad Creation

Amazon approached AI advertising differently. Amazon Ads expanded its AI Video Generator tool into India and seven other markets. Advertisers can now upload product images or use detail pages and generate six formatted video ads within minutes, at no additional cost.  It handles multi-scene storytelling, background music, text animation, and even a summary of existing clips.

Advertisers can use Amazon’s AI Video Generator to create eight-second “low-motion” videos based on their product. The tool is designed for smaller retailers unable to afford full-scale productions. Around 50% of products promoted with AI video ads are advertised for the first time, giving small businesses access to creative formats they could not afford before.

That said, the tool has limitations. Templates offer less creative flexibility than fully custom productions, and subtle elements like emotional depth, brand nuance, and bespoke storytelling are harder to achieve.

At its very core, Amazon’s tool for AI content creation is quick and inexpensive, but it still fails to match the finesse that human-directed video has.

Coign: AI Video Generation Goes Mainstream

Coign, a credit card brand, made history with the first fully AI-generated national TV commercial in financial services. The 30-second spot featured everyday Americans talking about responsible spending and core financial values. It was completed in half a day at less than 1% of the cost of a traditional production.

The ad proved that AI isn’t just for edgy brands; even conservative financial services can run prime-time campaigns built entirely by algorithms. But it wasn’t flawless. Critics pointed out that the conversation sounded mechanical, the time was not right, and the characters’ reactions did not match the natural flow of human actors. It got the job done, but it didn’t feel entirely alive.

What this shows is that yes, AI can cut costs and speed production, but it does not quite yet understand the subtleties, emotions and rhythms that humans bring to storytelling. Efficiency is powerful, but creativity still needs a human hand.

The Pattern Problem: Predictability in AI-Powered Advertising

When you step back and look at the examples, a clear trajectory appears. Coca‑Cola used AI in advertising to spark imagination and reinvent its iconic campaign visuals. Amazon deployed an AI advertising tool to optimise video ad creation with scale and cost‑efficiency. Coign showed that even traditionally conservative sectors can adopt generative AI for full‑scale video ad creation.

These are different uses of AI tools in advertising, yet they point to the same underlying trend: generative AI video ads and AI advertising tools are no longer novel experiments. They are becoming standard practice in the marketing industry.

But there is a caveat. Once brands rely too heavily on these AI video ad production tools, the results start to mirror each other. Their visual design, story structure and emotional storytelling risk converging into predictable patterns. Using AI in advertising is powerful, but it cannot replace the subtlety, cultural insight and human intuition that make an advertisement truly resonant.

What Agencies Should Take Away About AI in Advertising

Agencies cannot simply react by adopting AI advertising tools. The real question they need to ask is: what can we do that generative AI cannot?

In short, AI can do a lot of things such as: producing content, predicting outcomes, and optimizing campaigns on a large scale. For example, it can easily create AI video ads and execute AI-powered campaigns, but it doesn’t have the ability to understand culture, recognize nuance or even feel the timing. It cannot even come close to instinct, intuition, or emotional resonance. That human layer is where strategy and creativity become meaningful.

But you know what will differentiate digital marketing agencies going forward? It won’t be faster production or clever automation, but insight, depth, and emotional impact. AI in marketing is liberating; it can handle repetition, scale, and content creation quickly. But it also exposes lazy thinking.

The agencies that will win in the future will embrace AI as a way to augment human insight, not replace it. Creativity propelled by judgment, cultural sensitivity and instinctual storytelling will allow campaigns to continue to stand out.

Companies will continue to experiment, and AI tools will keep developing. The temptation of using efficiency as the main driver of the story will be there, but those agencies that are successful will still rely on human judgment, subtlety, and soul. AI is a great tool to make advertising better; however, it is still the human touch that makes the work truly memorable.