When More AI Means More Content, How Do Advertisers Find Quality?
This thought leadership was first published on NewDigitalAge.
$500 billion. That is the projected global investment in AI in 2026 alone, a scale comparable to rebuilding the entire US interstate highway system from scratch (Source).
One of the most visible effects of the increase of AI in media is content. Text, images, audio, and video can now be generated faster and cheaper than ever before. AI has dramatically increased the volume of content available across the open web, social platforms, and digital media environments.
This explosion in content isn’t inherently negative. But it introduces a structural challenge for advertisers: as content supply grows, identifying quality and consistently aligning ads with it becomes increasingly difficult.
When content explodes, quality signals weaken
For advertisers and agencies, the issue isn’t simply volume. It’s that the signals used to distinguish high-quality environments from low-quality ones become diluted as supply grows. As AI lowers the barrier to content creation and distribution, new sites, formats, and environments appear constantly.
Much of this inventory sits in a grey area: not fraudulent, not unsafe, but simply poor value. Ads technically run as intended, yet outcomes differ significantly depending on the surrounding content and format. Ads placed next to low-quality content don’t just underperform; they erode efficiency, weaken brand signals, and make optimisation more challenging.
Why current systems struggle
DSPs are optimised to maximise access to inventory, and their incentive structures reflect that. DSP revenue grows with spend and impression volume. As a result, quality controls like blocklists, brand-safety filters, and viewability thresholds tend to be reactive and coarse. At the same time, attention is one of the strongest indicators of media quality, but most buying platforms are not designed to dynamically optimise for it. This creates a structural tension; advertisers want their ads to appear in quality environments while buying systems are optimised to keep impressions flowing.
Research consistently shows how much environment and format matter. Adnami’s own attention research has found that the best format placed on the best domain can generate at least 70× more attention than the worst format on a low-quality domain, even when targeting and creative remain the same.
Industry data also highlights how difficult it remains to reliably filter quality at scale. According to eMarketer, more than 10% of all programmatic budgets are still estimated to land on MFA sites, representing spend that delivers little to no real value for advertisers.
Reducing complexity through automation
As the scale of the problem increases, agentic approaches become relevant and can remove much of the manual work of selecting and maintaining quality supply.
Agents can address quality by removing formats known to deliver low attention, excluding unverified or low-value domains, assessing attention and quality signals, and allowing only qualified inventory into buying environments at scale. Agents can also continuously analyse bidstream data in real time to continuously optimise for attention, making it possible to manage today’s media complexity at a speed and scale that manual processes alone cannot match.
At Adnami, this thinking has shaped our approach to agentic curation. Our Curation Optimiser Agents build on premium publisher relationships and Sonar attention metrics to create, package, analyse and optimise supply. The result is a simpler starting point for media buyers, where access to quality environments and attention signals is already in place before buying begins.
Final thoughts
As AI continues to increase content supply, the challenge for advertisers will not be accessing inventory, but identifying quality, aligning spend with it and optimising for attention. Agentic Curation is one possible response: reducing manual complexity, enabling optimisation for cost-efficient attention, and making quality a practical, scalable buying objective rather than an aspiration.
