Breaking Down the Walls: IAB Tech Lab’s Blueprint for Containerized, Agent-Powered Advertising

The programmatic advertising industry has reached an inflection point. As the digital advertising ecosystem operates as a distributed system allowing multi-party participation across various business activities, the need for a standardized approach to integrating new services has become critical. The new Agentic RTB Framework (ARTF) v1.0, now in Public Comment, addresses this need by providing a common specification that enables innovation while maintaining the performance, security, and operational requirements essential to real-time bidding.

Why does this matter?

The Agentic RTB Framework provides comprehensive specifications and guidance for deploying agents using containers in real-time programmatic and digital advertising use cases. The model leverages containers deployed into the infrastructure of a host to enable delegation of critical aspects of bidstream processing to service agents in a consistent manner, with minimal cost, latency, and operational impacts.

Adopting the framework will

Reduce Integration Overhead: Service providers can package their offering once and deploy it to any standard-compliant platform, allowing them to focus on their unique value proposition while offloading operational concerns and scaling to host platforms. This dramatically reduces the time and cost associated with integrating new capabilities across multiple partners.
Maintain Control and Security: Host platforms maintain control of data and SLAs and can provide greater access to data and more interaction opportunities to service agents without concerns about leakage, misappropriation, or latency. This addresses one of the most significant barriers to deeper integration between programmatic partners.
Enable Rapid Market Adaptation: Platforms can adapt quickly to changing market demands by simply adding, updating, and removing components as needed, all while maintaining control of operational costs and requirements without incurring significant integration overhead. In an industry where competitive advantage often depends on speed to market, this flexibility is invaluable.
Support a Composable Architecture: The framework enables platforms to compose bid processing pipelines configured specifically for their target use cases, creating new possibilities for innovation that weren’t feasible with traditional integration approaches.

Framework Design Principles

The framework is built on five critical requirements that ensure both practical adoption and technical excellence:

Bidstream Integration: Agents must participate in the core bidstream, focused on entities transacting in real-time
Intent-Driven Actions: Each agent must declare specific intents and any changes to auctions, allowing orchestrating entities to accept or reject changes
Enterprise-Grade Deployment: Agents must be structured as OCI-compliant containers manageable via Kubernetes, Docker Compose, or cloud-based systems like Amazon’s Elastic Container Service
Performance Excellence: Agents must communicate via high-performance protocols such as gRPC or MCP, written in efficient languages like Rust, Go, or Java
Security by Design: Agents operate with no external network access—all network ingress and egress is prohibited except for service communications with the orchestrating entity. This architecture ensures that sensitive bidstream data remains protected while still enabling powerful functionality.

Unlimited Use Cases

The framework intentionally avoids constraining implementation to specific use cases. Instead, it provides a general-purpose foundation that can support any product requiring multiple parties to communicate during the bidstream process using high-performance containers.

Example applications include:

User data collaboration: Agent (or AI) user data (identity, context and reinforcement signals) collaboration service to understand audiences that can process user signals with minimum latency, such as IAB Tech Lab’s open-source User Context Protocol (UCP)
Deal Activation and Curation: Dynamic deal management and optimization
Audience Segmentation: Real-time cohort activation and enrichment
Fraud Detection: Pre-impression fraud analysis and filtering
Bid Optimization: Real-time bid valuation and adjustment
Supply Path Optimization: Intelligent auction routing and prioritization

This list is incomplete by design, because part of the goal of this specification is to allow new use cases to be integrated into the bid stream via the agentic framework. The ARTFs flexibility ensures it can support use cases that haven’t even been imagined yet.

High Performance at Scale

Performance is achieved through multiple architectural choices:

Container-Based Deployment: Containers are “dropped in” to host networks and communicate through high-performance messaging mechanisms built on gRPC, ORTB Patch, and MCP. This approach ensures low to sub-millisecond response times compatible with real-time bidding requirements.

Protocol Efficiency: gRPC uses protobuf serialization, which is considerably more space and time efficient than JSON, provides easy validation with built-in rejection of invalid payloads, and offers a better model for agent-based communication than state transitions.

Semantic Addressing: Mutation paths use semantic references derived from OpenRTB concepts rather than explicit paths, allowing the spec to express mutations in terms of auction semantics rather than data layout. This flexibility enables orchestrators to optimize internal data structures without breaking agent integrations.

Open Standards and Interoperability

The specification leverages open protocols and industry standards throughout, as the interoperability is top of mind for all stakeholders to enable speed of adoption. Some of the open protocols included are:

OCI-Compliant Containers: Full compatibility with Docker, Kubernetes, and cloud container services
Protobuf Serialization: Open messaging protocol with schema definitions available in the framework’s GitHub repository
Standard Health Checks: Industry standards health and readiness probes
Open Telemetry: Built-in support for metrics and distributed tracing

The Future of Programmatic Infrastructure

The specification describes a standard interface with which containers can be managed using AI agents, enabling sub-millisecond real-time bidding operations driven by agentic systems. While the current focus is on systematic agentic integration (service-to-service), autonomic agentic functionality (model-to-service) is also envisioned as part of this specification as the integrating technology matures.

This forward-looking approach positions the industry to take advantage of rapidly evolving AI and machine learning technologies while maintaining the performance and reliability requirements of programmatic advertising.

IAB Tech Lab continues to work on specifications to empower communication between companies using emerging technologies. There are a lot of new products being built on top of agentic systems. With the Agentic RTB Framework a next generation set of use cases can be unlocked and standardized across multiple companies. Some of the most innovative companies participated in this collaborative effort including Index Exchange, OpenX, The Trade Desk, & Chalice.

Join the Container Project Task Force to shape the future of agentic standards today!