IAB Tech Lab's AAMP 3.0 vs. AdCP: Who Actually Controls Agent-to-Agent Ad Buying Now?

What does AAMP 3.0 actually standardize?

Based on IAB Tech Lab's own framing: the RFP-to-buy process specifically, meaning how an agent representing a buyer requests proposals from sellers, and how that negotiation resolves into an executed buy, in a standardized, interoperable way. That's a narrower and more specific scope than "the protocol for agentic advertising" broadly; it's targeting one particular workflow inside the larger agent-to-agent commerce picture.

IAB Tech Lab's own positioning, as reported, frames this as filling a gap AdCP doesn't fully address, implying AAMP 3.0 isn't simply a redundant competitor covering identical ground, but an attempt to standardize a piece of the workflow the existing protocol handles less completely. Independent, more neutral evaluation of that specific claim is still developing as of this writing, and it's worth tracking rather than accepting at face value from either side.

How is this different from AdCP?

AdCP, the Ad Context Protocol, is what this site has previously covered for agent-to-agent ad buying, with more direct platform backing from companies building agentic ad-buying products, OpenAI notably among them. AAMP 3.0 comes from IAB Tech Lab, the standards body with deep roots across the traditional programmatic advertising ecosystem, publishers, agencies, ad tech vendors who've operated under IAB standards for years.

That difference in origin is the more important story than any specific technical distinction at this early stage. AdCP represents newer, platform-native agentic-commerce infrastructure defining its own rules. AAMP 3.0 represents the establishment standards body asserting that agentic advertising should route through the same kind of open, multi-stakeholder standardization process that's governed programmatic advertising for years. Which philosophy wins, or whether both persist, serving different parts of the ecosystem, is the real open question, not just which technical spec is better engineered.

Why does a protocol-level split matter to a team running paid media day to day?

Because protocols determine what an execution layer can technically plug into, and a fragmented standards landscape means more integration work, not less, for longer than a single winning standard would require. If AAMP 3.0 and AdCP both gain real adoption and never converge, any platform or execution layer serious about agent-to-agent buying eventually needs to support both, which is real engineering cost passed down the chain, ultimately showing up as either a feature gap or a delay in whichever vendor you depend on for this capability.

The more immediate, practical implication: don't take a vendor's claim of "AdCP support" or "AAMP 3.0 support" as evidence of broad agentic-commerce readiness. As of September 2026, supporting one protocol says nothing about supporting the other, and the market hasn't told you yet which one, or whether both, will matter in twelve months.

Is this part of a bigger pattern in agentic commerce standards?

Yes, and naming that pattern is more useful than treating AAMP 3.0 vs. AdCP as an isolated rivalry. The broader agentic-commerce and advertising landscape in 2026 has multiple emerging protocols competing for different pieces of the same general problem: Google's AP2 for agent payments, OpenAI's own ACP for agentic checkout, AdCP for agent-to-agent ad buying, and now AAMP 3.0 entering the ad-buying-specifically space that AdCP already occupied. None of these are formally unified, and each is backed by a different part of the industry with a different incentive in how the eventual standard gets written.

The historical pattern in technical standards wars, multiple competing specs, eventual consolidation around one dominant standard or a negotiated interoperability layer, is the likely eventual shape here too. The honest, current answer is that it hasn't happened yet, and betting heavily on any single protocol today is a bet on which side of that eventual consolidation you'll land on.

What should a team actually do while this is unresolved?

Track adoption signals, not announcements. A protocol getting announced is a much weaker signal than a protocol getting adopted by multiple independent platforms and vendors. Watch who actually implements AAMP 3.0 over the next two quarters before treating it as established.

Avoid deep, protocol-specific lock-in where you can. If your execution layer or ad-buying stack has a choice between a flexible integration approach and a deep, protocol-specific build against one standard, the flexible approach is the safer bet during an active standards dispute, the switching cost of having bet wrong compounds the longer you've built against the losing standard.

Ask your vendors directly which protocols they support today, and which they're watching. A vendor with a clear, honest answer about its current protocol support and its posture on the unresolved ones is a more trustworthy partner in this specific moment than one that claims full agentic-commerce readiness without naming which standards that actually covers.