AI in Advertising: Big Promises Clash with a Sobering Reality

NEW YORK, NY – January 22, 2026 – A significant disconnect is emerging in the world of television advertising: while a vast majority of advertisers believe artificial intelligence is a transformative force, most have yet to see it deliver meaningful business results. A new report from Comcast Advertising highlights this growing gap between hype and reality, revealing an industry grappling with the practical challenges of turning AI's immense potential into proven returns.

The annual report, which surveyed 216 senior brand and agency leaders in late 2025, found that nearly eight in ten advertisers (77%) believe AI is fundamentally changing how TV advertising is bought and sold. They credit the technology with making the process more efficient (60%) and easier to manage (37%). Yet, despite this optimism, a sobering 61% confess they have not seen any meaningful business results from their AI investments, underscoring a critical juncture for the ad tech landscape.

“When done right, AI can support making TV advertising simpler, smarter, more efficient, and even more accessible,” said James Rooke, President, Comcast Advertising, in the report's release. He noted, however, that the industry is still in the "very early stages of realizing the benefits," and that less than a third of advertisers fully trust AI with advertising-related tasks. The central challenge, he added, "is to translate AI’s potential into practical innovation that goes beyond just buzz to deliver real business value.”

The Great Expectation Gap

The chasm between high expectations and current outcomes is a defining feature of the AI adoption curve in advertising. The Comcast report’s findings are not an anomaly; they reflect a broader industry sentiment. Nielsen's 2025 Annual Marketing Report found that 74% of marketers considered AI "critically important" for their success, yet other analyses, like Deloitte's 2026 State of AI report, show that realized benefits are still heavily skewed towards improving efficiency rather than directly boosting revenue.

This gap suggests that while advertisers have successfully integrated AI to streamline workflows, the strategic implementation required to drive measurable growth remains a significant hurdle. The industry is moving past the initial phase of excitement and entering a more challenging period of proving ROI, a necessary step to secure continued investment and justify the technology's transformative billing.

The AI Proving Ground: Where It Works and Where It's Headed

For now, advertisers are overwhelmingly deploying AI for foundational, efficiency-driving tasks. According to the Comcast data, the most common applications are identifying and segmenting audiences and analyzing consumer behavior, with 82% of respondents using AI for these purposes. Close behind is automated data collection and integration, cited by 80% of those surveyed.

These applications represent the low-hanging fruit of AI adoption, automating complex data-sifting tasks that were once manual and time-consuming. Platforms from major players like Google, Meta, and The Trade Desk have made it easier for marketers to leverage machine learning for automated bidding and audience targeting. However, these efficiency gains have not consistently translated into the bottom-line impact that executives demand.

The industry's gaze is now fixed on a more complex prize: measurement and attribution. The report reveals this is the top area (30%) where advertisers hope AI will make a significant impact in 2026. Proving that a specific ad on a specific platform led to a sale has long been the holy grail of advertising. Marketers are betting that AI, with its ability to analyze countless data points across a fragmented media landscape, can finally provide the clear, closed-loop measurement needed to justify massive ad spends.

The New Creative Canvas: From Robot Brainstorms to Digital Actors

AI's influence is also seeping into the creative side of the business, though adoption remains cautious. The Comcast survey shows advertisers are most comfortable using AI for initial brainstorming (41%) and generating multiple versions of a creative concept (35%). These tools allow teams to rapidly iterate on copy and visuals, testing what resonates with different audience segments.

However, only one in five advertisers (20%) report using AI for full ad production. This hesitation highlights the ongoing tension between AI's power to scale content and the perceived need for human-led creative strategy and brand stewardship.

Despite this caution, pioneering brands are offering a glimpse of what's possible. Nike's "Never Done Evolving" campaign used machine learning to create a stunning video of Serena Williams playing against AI-generated versions of her younger self, earning over 100 million views. Similarly, Coca-Cola's "Create Real Magic" platform invited artists to co-create brand artwork using DALL-E and GPT-4, with AI guardrails ensuring brand consistency across thousands of unique submissions. These high-profile successes demonstrate that when paired with strong human creative direction, AI can produce emotionally resonant and highly effective advertising.

Navigating the Hurdles: Why 61% Are Still Waiting for a Payoff

The gap between AI's potential and its current impact can be attributed to a confluence of practical and ethical challenges. The primary obstacle is not the technology itself, but the human and organizational infrastructure surrounding it. A 2026 report from Deloitte found that insufficient worker skills were a top barrier to AI adoption for 53% of business leaders, a sentiment that resonates deeply within the specialized world of advertising.

Data quality is another major stumbling block. AI models are only as good as the data they are trained on, and many organizations are struggling with siloed, incomplete, or inaccurate datasets. Feeding flawed data into a sophisticated algorithm doesn't produce insight; it amplifies errors at an unprecedented scale and speed.

Furthermore, the very complexity that makes AI powerful also makes its ROI difficult to prove, creating a chicken-and-egg problem for budget holders. Without robust frameworks to test and measure AI's incremental impact, securing further investment becomes a challenge. This is compounded by a market flooded with vendors whose "AI-powered" claims often outpace their actual capabilities, forcing marketers to separate genuine innovation from marketing hype.

Looming over these practical issues are significant ethical and legal questions. Navigating a patchwork of data privacy regulations like GDPR and CCPA is a constant concern. The risk of algorithmic bias, where AI models perpetuate or even amplify societal biases present in training data, poses a serious threat to brand reputation. As AI-generated content becomes more sophisticated, issues of transparency and copyright—particularly whether creators should be compensated when their work is used to train AI models—are becoming central legal battlegrounds for the industry. Until these hurdles are addressed, the journey from AI's promise to widespread, profitable implementation will remain a work in progress.