Akamai's Agentic Security Framework: The Trust Layer for Autonomous AI Commerce

The Six Pillars of Akamai’s Agentic Security Framework

1. Verified Identity and Human Attribution
   This pillar establishes a verifiable link between an autonomous AI agent and the human user who authorized it. The goal is to ensure that every action taken by an agent can be traced back to a real, consenting individual . This is achieved in part through partnerships with Visa and Experian, which contribute identity-validation and payment-verification capabilities .

2. Unparalleled Visibility Across All Traffic Sources
   Many websites today cannot reliably distinguish between a legitimate AI agent gathering information for a user and a malicious bot scraping data. This pillar provides deep observability into traffic patterns, enabling organizations to identify, classify, and understand agentic traffic flows for the first time .

3. Adaptive Trust Analysis
   Rather than a binary allow-or-block decision, the framework applies a sliding-scale model. An agent’s trustworthiness is evaluated in real time by combining identity signals with behavioral analysis. A request might be fully allowed, rate-limited, challenged, or blocked based on a continuous trust score, creating a more nuanced and efficient security posture .

4. Edge-Powered Decision-Making
   All enforcement decisions are executed at Akamai’s globally distributed edge network. By processing trust signals and terminating untrusted or anomalous requests at the edge before they reach a customer’s origin server, the framework helps protect backend infrastructure from both volumetric and targeted attacks without impacting performance for legitimate traffic .

5. Monetization Enablement
   Content publishers can use this pillar to charge AI agents on a pay-per-request basis. Built with partners like TollBit, a licensing and payments platform, and Skyfire, an agent-identity and payment infrastructure provider, the system is intended to enforce paid access terms at the edge. This converts previously uncontrolled agent access into a permissioned, monetizable channel .

6. User-Centric Analysis
   Behavioral and contextual analytics are tied directly to the authorized user behind each agent. This allows organizations to apply existing human-user risk models and usage policies to the actions of their corresponding agents .

The Know Your Agent (KYA) Protocol: Binding Humans to Bots

The central mechanism underpinning the framework is the Know Your Agent (KYA) protocol. It is designed to answer three fundamental questions for every request: “Who is the agent, who authorized it, and what is it allowed to do?” . The protocol works by creating a cryptographically verifiable chain that connects a human user, their AI agent, and a specific transaction.

This is achieved through a strategic partner ecosystem:

• Visa: Akamai integrates its edge security with Visa’s Trusted Agent Protocol (TAP), which provides a framework for authenticating AI agents that use Visa credentials for payments. Through this partnership, Akamai adds behavioral and network intelligence to the payment credential verification that Visa provides .
• Experian: Akamai plugs into the Experian Agent Trust™ framework. This framework establishes a system of human-to-agent binding, creating a verifiable and persistent link between a consumer’s identity and their agent. Akamai’s edge capabilities strengthen this by enforcing security policy and controlling bot activity in real time .
• Skyfire: This partner provides payment and agent-identity infrastructure that enables agents to authenticate, operate within policy, and access global payment rails. The integration brings this trust layer to the edge without requiring enterprises to re-architect their systems .

The practical result is that a trust decision can be made before a single request reaches an origin server. As one partner described it, Akamai’s edge enforcement plugs into Experian’s human-to-agent binding to add real-time risk scoring, allowing a merchant to decide whether to trust an agent-driven transaction pre-emptively .

Spectrum-of-Trust Evaluation at the Edge

A core design principle of the framework is its adaptive trust analysis, which moves beyond the traditional binary firewall logic. Agent traffic is evaluated on a sliding scale, with a trust score assembled from multiple real-time signals: the verified identity from the KYA protocol, the agent’s behavioral patterns, and the context of the specific request . Enforcement actions—ranging from allowing and monitoring to limiting, challenging, or blocking—are executed instantly at the distributed network edge . This is intended to stop untrusted automation early while enabling seamless, low-latency interaction for verified agents.

Enterprise Identity Provider Integration

The framework is designed to fit into an enterprise’s existing security architecture, not replace it. It is built to integrate with common identity providers, allowing organizations to extend their current authentication and authorization policies to agent-driven traffic . Although Auth0 and Ping Identity are not named in the available sources on this specific framework, Akamai’s description of the pillar indicates a goal of bringing standard identity infrastructure into agent trust decisions, so an agent’s access can be governed by the same IAM context already used for human users .

A New Monetization Model: Pay-Per-Request for AI Agents

Perhaps the most commercially transformative pillar is Monetization Enablement. For content publishers, AI agents represent a double-edged sword: they drive traffic but often bypass ads and paywalls. This pillar offers a direct economic answer. By partnering with TollBit, a licensing and payments platform, and Skyfire, Akamai enables a publisher to set access rules that charge agents on a per-request or usage basis .

In this model, a publisher can define terms for how an AI agent is allowed to consume its content. Akamai’s edge network enforces this policy—granting access only to authenticated, paying agents—while payment partners handle the usage tracking, billing, and settlement . The vision is to turn the tide of unmonetized agent scraping into a new, high-volume revenue channel where access is permissioned and paid for.

Together, Akamai’s six pillars and its partner network represent a systematic bet on a future where much of the web’s traffic is non-human. By shifting the security model from blocking to managing and monetizing, the framework aims to provide the trust layer that autonomous AI commerce cannot scale without .