With the rise of autonomous AI agents, the advertising industry must work together to ensure best practice and shared progress, explains David Rudnick of LG Ad Solutions.
The advertising industry must rise to the challenge of establishing a unified framework for the implementation of AI agents, says Rudnick
The advertising industry is standing at the threshold of a transformation unlike any we’ve seen before. Not since the rise of programmatic buying or the proliferation of connected TV (CTV) has a new paradigm held such potential to fundamentally reshape the ad ecosystem. But this time, it is not just about efficiency, reach, or real-time bidding. It is about autonomy.
This pivotal moment is driven by the emergence of AI agents and model context protocol (MCP): specialized, intelligent systems trained to perform specific tasks throughout the advertising workflow. These agents represent a leap forward from conventional automation. Rather than speeding up repetitive tasks, AI agents learn, adapt, and act. They are capable of understanding complex data environments, making decisions in context, and collaborating across systems.
AI agents will become foundational to the future of digital advertising. They will not only make campaigns more efficient but enable new levels of responsiveness, personalization, and intelligence that marketers have long been promised but rarely delivered at scale.
Why agents, and why now?
To understand the significance of AI agents, it’s important to consider the current state of advertising. Most brands and platforms are already overwhelmed by data. Every campaign, impression, and conversion generates signals – billions of them. But making sense of that data, connecting it across platforms, and turning it into action still requires considerable manual effort or brittle workflows with limited scalability.
Traditional automation has helped. We have built tools that schedule campaigns, generate reports, and optimize bidding strategies. But these tools are largely rules-based. They do not reason. They do not collaborate across systems. They do not adapt dynamically to the nuance of real-world constraints, privacy requirements, or shifting business goals.
AI agents are different. They operate more like virtual teammates. They’re capable of digesting massive amounts of structured and unstructured data. They can reason based on defined parameters and execute actions across disparate environments. For example, an agent could analyze media mix performance, identify underperforming segments, reallocate spend, and generate reporting – all in near real-time – without human intervention.
This jump from automation to autonomy is not just a technical milestone. It is a shift in how the advertising industry can operate. And it is a move that I believe is essential to unlocking the next wave of growth and innovation.
Building the right infrastructure
Of course, none of this works without the right infrastructure in place. To perform to their full potential, AI agents require access to clean, centralized data, powerful models, and a secure deployment climate.
Want to go deeper? Ask The Drum
The challenge here is for the industry as a whole to step up in developing a robust and unified stack to work from. There are plenty of market options available when it comes to scalable platforms that can deliver the warehousing, flexibility, and integrative grunt needed to optimize AI agents in advertising.
Yet, the unique nature of the digital ecosystem means we must go one step further. The ideal next move involves the development of cloud-based solutions that support both the deployment of agents and their capacity to understand context.
Because the best kind of system in an agent-driven era of advertising will be able to adapt seamlessly across different habitats. This might mean operating within the workflows of a walled garden, or adhering to permissions set by diverse actors – whether that’s clients or internal objectives. It may also involve maximizing data availability on an open web exchange or in a private data clean room.
Whatever the context, we need a uniform system that’s sophisticated enough to first share enterprise-grade data, and then enable AI agents to understand and switch modes according to a huge range of digital settings and protocols. In addition, this structure needs to be accessible enough for brands of all sizes and backgrounds to be able to use with ease.
An agentic consortium
A standard framework that makes AI agents interoperable is the dream – but how do we get there? The answer lies in a sector-wide consortium designed to harness the collective voice of digital advertising in AI. Advertising platforms, media companies, data providers, and tech partners must come together in a group effort to set boundaries within agentic AI’s thorniest issues (eg accountability, compliance, and security measures).
The idea is that every party involved is free to build its own suite of AI agents, tailored to particular brand preferences and goals. Yet the wider body will help put vital guardrails in place. Just as the industry once coalesced around standards like VAST, OpenRTB, and SKAdNetwork, it’s now time to do the same for AI agents.
We are all too aware of the dangers at play otherwise. Fragmentation, lack of transparency, incompatible tools, and privacy risks have followed nearly every wave of digital transformation in the past 20 years. If we want AI agents to truly thrive and scale, we need to get ahead of that – keeping innovation in line with key standardization efforts.
Tomorrow’s marketplace
Alongside the consortium, the industry would benefit from the development of a tailored marketplace for agentic technologies – a central hub where members can create, trade, and maintain AI agents across the digital ad ecosystem. One company might contribute an agent for forecasting, another for audience modeling, and another for financial reconciliation.
Having this more tailored marketplace would make it easy for companies to adopt proven agents or adapt them based on their proprietary data and goals. It provides a new model of collaboration: a modular, community-driven approach that acts as a rallying call to the industry’s brightest pioneers.
If we know anything from history’s biggest game-changers in advertising, it’s that they rarely work in isolation. Whether it’s the rise of programmatic media or video streaming, each new phase of innovation demands teamwork to manage both the scale of opportunity and the risks involved.
Let’s shape this new agentic chapter with an alliance that values openness, intelligence, and shared progress. From infrastructure and standards to controls and creative sharing, let’s build the future together.