Could AI Undermine YouTube’s Mission To Grab More TV Ad Dollars?

Artificial intelligence is rapidly reshaping how brands are approaching their marketing strategies, and that includes video production. The ease at which AI enables anyone to churn out a video in minutes is breathtaking, and likely a bit of a siren's call for overworked marketers on tight budgets. 

But this ease has translated into an avalanche of low-quality AI content. In fact, as CNBC reported, YouTube CEO Neal Mohan said “reducing ‘AI slop’ and detecting deepfakes are priorities” for this year. 

That certainly makes sense, as YouTube increasingly tries to snag a bigger piece of advertisers’ traditional TV budgets. Google's president of Americas and global partners, Sean Downey, told Business Insider, “In this era of entertainment, YouTube is a brand's best bet for staying relevant. YouTube has original content viewers love, the trusted creators who are driving culture forward, and the innovative ad solutions that deliver results advertisers can't find elsewhere.”

As platforms race to contain low-quality AI content and consumers grow more adept at spotting it, a clear warning sign is emerging for marketers: Efficiency gains mean little if trust erodes along the way.

New data from Animoto’s 2026 State of Video Report suggests consumers are far more perceptive about AI-generated video than some marketers assume. Over 82% say they believe they can identify AI-created videos, and 36% say that seeing AI video content lowers their trust in a brand. The most common red flags include stiff or robotic movement, unnatural voiceovers and content that lacks emotional depth.

Here’s a real-world example: One of Coca-Cola’s 2025 holiday ads, which was an AI-generated remake of its 1995 “Holidays Are Coming” commercial. Viewers were quick to point out multiple AI issues, such as the number of wheels on a truck changing from scene to scene. Many people, especially those in creative industries, expressed displeasure or poked fun at the company in online forums such as Reddit and in YouTube comments. 

This matters because video has become one of the most powerful drivers of consumer decision-making — the vast majority say video is both the most memorable format and their preferred way to learn about brands, and 83% report making a purchase after watching a brand video. 

Meanwhile, marketers are moving full speed ahead. Animoto’s report shows that 84% of marketers now use AI in video creation, with most relying on it regularly. For the most part, AI isn’t replacing creative teams (yet), instead it’s being used to accelerate workflows by brainstorming ideas, writing scripts and editing. For many marketers, generating ideas remains the biggest bottleneck, and AI has become a go-to solution.

But even as adoption surges, marketers themselves recognize the limits. Nearly all say maintaining creative control, brand personality and the ability to edit AI-generated content is essential. 

This gap between how brands use AI and how consumers perceive it is where risk creeps in, especially as platforms like YouTube work to curb AI slop. Brands that churn out generic, emotionally flat video may find themselves filtered out by both algorithms and audiences, leading to waste in ad spend, especially if TV budgets start being diverted to YouTube. Consumers still want short, human, authentic stories — preferably featuring real people — and they are quick to disengage when content feels automated.

For brands not paying attention, the danger isn’t missing the AI wave. It’s going all-in without acknowledging consumer sentiment, and reversing the trust that video was supposed to build all along.

Eleanor Semeraro

Eleanor is an entertainment analyst and marketing strategist with a passion for all things TV and social media. She’s a regular TV[R]EV contributor and consults for small businesses within the advertising and entertainment data analytics ecosystem.

Next
Next

Netflix's Ad Tier Is Still A Cypher, Linear Takes Another Hit