AI agents and so-called “super signal aggregators” are being framed as the saviors of premium publishers and advertisers who prioritize high-quality media.
But these technologies could perpetuate programmatic’s worst practices and push premium media brands into a Commoditization 2.0 era.
Commoditization 1.0 occurred when quality media brands legitimized the programmatic open marketplace by participating in it. That choice diluted their value, exposed their brands to reputational risks and ultimately harmed their long-term revenue potential.
Even now, not all media owners fully recognize that many ad-funded models are not only unsustainable but also work against their interests. And they could be vulnerable to further commoditization as AI slop overtakes the internet and AI agents assume control of programmatic buying.
Further commoditization on the publisher side would also negatively affect advertisers with more sophisticated programmatic strategies than spread betting on ad inventory and hunting for low CPMs.
But, by avoiding the risks inherent in further automation of a flawed marketplace, we may be able to realize the potential of these emerging technologies – and build a more sustainable open web.
The promise of AI as a fix
The new narrative is that AI can fix the old open marketplace by adding a smart layer on top of it – that algorithms can somehow sift through the swamp to identify quality media at scale.
Perhaps AI can deliver on that promise. I remain open-minded about that possibility. But even if the open marketplace becomes “better,” it still isn’t what premium advertisers and media brands need.
If agentic AI plays the role of a smarter sifter, finding occasional moments of quality across millions of sites, fine. That’s helpful for tier 2 advertising strategies that focus on the bottom 70% of publishers. But the top 30% of media owners should not be the first to blink again, seduced by the promise of another gravity-defying dream of incremental revenue.
Conflicting business models demand tailored answers
The digital advertising ecosystem has long been built on conflicting business models and, unless these models are fundamentally changed, the rise of AI risks compounding the contradictions inherent in the marketplace rather than ushering in a new era of progress. We’ll be doomed to automate the same old mistakes.
A large part of the industry, particularly ad tech, still clings to a model that prioritizes reach above all else at the inventory entry point. Their business depends on it: Their role is to find diamonds in the sewage, so they need the sewage.
Meanwhile, advertisers and quality media owners often find themselves pulled toward that same flawed logic. Publishers try to prove they’re diamonds, and buyers get used to sifting through the sewer.
Some industry factions behave more like sects than professionals with a holistic understanding of their own ecosystem. The “ad tech for ad tech’s sake” crowd will dismiss these concerns about automating media quality decisions as Luddite nonsense. They have already drunk the ad tech Kool-Aid.
Simultaneously, some privacy advocates may uncritically embrace any criticism of AI tech, as long as it’s seen as opposing data-driven advertising. But today’s world demands more nuance and maturity.
The evolution of premium media owners
First, we must acknowledge that premium media owners and their business models have already evolved.
Audience-centricity is no longer optional; it is the backbone of any successful diversification strategy. The value of a media brand doesn’t lie solely in its content, data, trust or distribution but in how those assets are woven together into an environment that attracts, engages and retains audiences.
Those assets should not be decoupled and undersold like bags of popcorn at the closing of a funfair on Sunday night. If advertising is to be called premium, it must live within a premium environment.
Quality over liquidity
Since the EU implemented GDPR, much of the real progress in digital advertising has come not from technology alone but from quality media owners deliberately distinguishing themselves from the long tail. They renewed their focus on the sum of their assets: exclusive environments built on trust, relevance and direct audience relationships.
That’s still the path forward in the age of agentic AI.
Premium media environments cannot be disaggregated and rebuilt by algorithms scanning the open web. Their value, and the price attached to it, is in the blend of content, context and connection, not in their individual ingredients.
Until AI can accurately account for a publisher’s holistic value; we can’t let it be the arbiter of media quality.
“The Sell Sider” is a column written by the sell side of the digital media community.
Follow Alessandro De Zanche and AdExchanger on LinkedIn.
For more articles featuring Alessandro De Zanche, click here.
