The advertising industry is full of noise about AI making buy-side and sell-side processes more efficient. That framing is convenient, but it misses a broader point. Speed and effectiveness are easy to celebrate. Power dynamics are harder to talk about.
Most conversations about AI in ad tech focus on fewer manual steps, more dynamic creative, quicker and more personalized optimization and leaner teams. Those gains certainly matter, but they do not fundamentally change who controls data, who makes decisions or who captures value. In many cases, automation simply allows existing power structures to operate more smoothly inside closed systems.
The more consequential shift begins when AI moves beyond execution support and starts expanding who can access data and act on it.
Access is the real disruption
When intelligence becomes accessible instead of centralized, the architecture itself starts to change. That change challenges long-held assumptions the ad tech stack has been built on, including the idea that insight and decision-making should remain tightly guarded and closed platforms are the fastest way to scale. Keeping pace with AI-driven change increasingly depends on how composable and adaptable systems are, not how tightly controlled they are.
Over the next few years, this will stop being a theoretical debate. As AI systems take on greater responsibility for planning, activation and optimization, business models built on control may weaken as more players gain access to data and the ability to apply it intelligently.
Advertising technology has historically thrived on scarcity. High-quality data, advanced analytics and real-time decision-making were concentrated among the largest platforms and the most technical organizations. Brands, agencies and publishers downstream were left responding to insights they couldn’t fully interrogate or influence. Smaller brands and publishers were often excluded, except inside walled gardens.
AI has the potential to change this balance if it is applied differently. Used narrowly, AI accelerates familiar dynamics: campaigns move faster, optimizations happen more frequently and reporting improves. Transparency and collaboration do not necessarily follow, however, and the rich could just get richer. The stack may become more efficient but less open and effective for the masses.
From insight to agency
Agentic AI introduces a more disruptive path. When systems move from producing insights to planning and executing decisions, the constraints shift. Access to data, effective governance and clarity of objectives become more important than technical sophistication alone. More organizations will be able to query data directly, test assumptions and act in real time, constructing and composing their own solutions without relying on as much engineering or as many intermediaries.
This is where the conversation often becomes uncomfortable. When access to insight and execution increases, entire layers of the ecosystem built on mediation, opacity or control become harder to justify. It is no coincidence that automation tends to be celebrated more loudly than access; automation feels safe, while democratization forces a redistribution of influence.
When controls stops compounding
As access expands, the structure of the ad tech ecosystem will change in visible ways. Certain intermediaries will fade in importance. Decision-making will move closer to data owners at the ends of the supply chain. And composable components will increasingly become the connective tissue of modern marketing operations.
This pressure explains why consolidation is accelerating across the industry. The organizations that thrive will be the ones that support collaborative but governed data access, broader execution and shared measurement, rather than forcing every participant into one walled garden or another. In a more open, AI-driven ecosystem, influence comes from helping others use data effectively and responsibly, not hoarding it. Efficiency gains are incremental; changes in access are structural.
The result: Platforms that enable shared access and accountable decision-making will grow in relevance, while those that depend on friction or opacity will struggle to justify their roles.
The evolution of advertising over the next few years will not be defined by how efficiently AI can execute decisions; it will be defined by who and what gets to be involved in making them.
“Data-Driven Thinking” is written by members of the media community and contains fresh ideas on the digital revolution in media.
Follow Snowflake and AdExchanger on LinkedIn.
For more articles featuring Dennis Buchheim, click here.
