Observed Signal · Nov 11, 2026 · Technical Release · Source: Adzine · Impact: 3/5 · Sentiment: Positive
Will Agencies Bring Their Own Algorithms to Media Buying?
In an interview with ADZINE, César Masumbuku, Senior Advertiser Solutions at PubMatic, discusses the IAB Tech Lab's Agentic Real Time Framework (ARTF), which standardizes how buyer logic is deployed and executed in real-time ad auctions. ARTF allows models to be containerized and run directly within a host platform's infrastructure, such as an SSP or DSP, without external calls that waste milliseconds. This enables agencies to bring their own algorithms into the auction, but the host remains the gatekeeper, deciding which data is shared and which changes are accepted. Masumbuku explains that while results may vary across hosts, the core decision framework remains portable. He also introduces PubMatic's Decision Fabric, a containerization layer built on Agentic OS that executes partner decision models natively within the programmatic supply path, currently used for curation and traffic shaping. The framework aims to give agencies more choice and control over their media buying logic while leveraging platform infrastructure and signals.
This article discusses the emerging ARTF standard for running agency algorithms within ad platforms, which could reshape media buying and increase agency control. It is a significant industry advancement by the IAB Tech Lab and PubMatic.
Track PubMatic Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your Polaris7 workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- ARTF (Agentic Real Time Framework) is an IAB Tech Lab standard, version 1.0 completed in August 2026.
- ARTF allows buyer models to be containerized and run within host platforms like SSPs, exchanges, or DSPs.
- PubMatic offers Decision Fabric, a containerization layer on Agentic OS for running partner decision models.
- Current use cases for Decision Fabric containers include curation and traffic shaping.
- The host platform retains control over data sharing and execution of model suggestions.
Connected Companies & Entities
“César Masumbuku, Senior Advertiser Solutions at PubMatic, discusses ARTF and Decision Fabric.”
“ARTF is a standard developed by the IAB Tech Lab.”
“The host remains the gatekeeper, quoting IAB UK.”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
Unlocking Ad Tech Efficiency with New ARTF Standard
The IAB Tech Lab published a practical five-step guide to ARTF (Agentic Real-Time Framework), a new standard that lets vendors run decisioning logic inside programmatic auctions as sandboxed containers. ARTF operates on top of OpenRTB: containers receive bid requests, return OpenRTB Patch responses proposing intent‑tagged mutations, and communicate via gRPC or MCP. Hosts (typically SSPs or exchanges) control inputs, cryptographically verify container images, and decide which proposed mutations to apply. The guide explains how companies should evaluate use cases, design low‑latency decisioning logic, package containers to the spec, pursue certification, and partner with host platforms for deployment. Index Exchange is named as an active participant in early ARTF discussions. The spec is public (artf.ai and on GitHub) and targets auction‑time use cases such as fraud detection, identity resolution, contextual enrichment and outcome‑based optimization.
Adform, Adsquare add AI-agent data signals to DSP
Adform and Adsquare jointly implemented the Agentic Real-Time Framework (ARTF) — part of the industry Agentic Advertising Management Protocol (AAMP) — inside the Adform Demand‑Side Platform (DSP). ARTF provides a standardized execution layer to enrich and process Bid Requests, enabling external AI agents and data vendors to surface signals (e.g., geographic and offline data from Adsquare) for real‑time media decisions without exporting platform data. The move aims to reduce bespoke integrations between data providers and DSPs and to accelerate agentic, multisignal programmatic buying. Adform had previously opened its DSP to AI agents via the Model Context Protocol (MCP); ARTF complements MCP by addressing operational processing of external signals within programmatic systems. Both companies position the implementation as an early step toward technical interoperability in agentic advertising, subject to broader market adoption.
Agentic Sidecar Architecture for Real-Time Programmatic Ads
IAB Tech Lab's Agentic Real-Time Framework (ARTF) v1.0 advocates a containerized "Agentic Sidecar" architecture to enable real-time, agent-driven programmatic advertising. The sidecar runs within a host platform's (DSP/SSP) virtual environment rather than calling remote cloud LLMs, addressing strict 300–500ms auction latency requirements. The specification recommends binary, high-performance communication via gRPC and Protocol Buffers (Protobuf) in place of JSON, claims up to an 80% reduction in bid request processing time, and describes atomic "Bidstream Mutations" executed in sub-millisecond windows. Governance is handled by a "Propose-and-Apply" model where agents submit intent-declared mutations (e.g., ADJUST_DEAL_FLOOR) for host approval. The sidecar is also positioned as a universal adapter for channels like DOOH and CTV by translating complex signals into OpenRTB patches. The article is authored by Gulab Patil, Founder and CEO of Lemma, and published 2026-04-30.
Receive all future signals for PubMatic & competitors in your briefing
Never miss material M&A deals, leadership changes, account wins, or product rollouts. Polaris7 monitors 20,000+ primary sources 24/7, delivering verified telemetry directly into your executive briefing.