Akamai has announced a strategic collaboration with Visa to bring stronger identity, user recognition, and security controls to agentic commerce.
Following this announcement, through the integration of Visa’s Trusted Agent Protocol with Akamai’s edge-based behavioral intelligence, user recognition, and bot and abuse protection, the partnership is expected to enable the companies to deliver the identity, authentication and fraud controls required to allow merchants welcome AI agents with commerce intent into their digital storefronts.
Strengthening the trust layer in agentic commerce
According to the official press release, as autonomous AI agents increasingly browse, compare, and purchase on behalf of consumers, merchants and businesses face a new array of challenges and difficulties. Through this development, companies will need to be able to differentiate this new type of legitimate automated traffic through the process of authenticating the agent, identifying the user interacting with it, and ensuring the interaction is safe and trusted. Without this trust layer, merchants risk losing control of personalisation, security, as well as the consumer relationship.
With this in mind, through the use of the Trusted Agent Protocol’s agent authentication framework and Akamai’s edge-based behavioral intelligence and user recognition, merchants and businesses will have the possibility to gain access to precise, real-time insight into AI agent activity before it touches sensitive systems. Moreover, this unified approach will allow them to confidently differentiate trusted AI agents from malicious bots, as well as unlock the potential of agentic commerce. At the same time, with Trusted Agent Protocol supported across Akamai Cloud, businesses will be enabled to operate at the speed and scale agentic commerce demands.
Furthermore, according to Akamai's 2025 Digital Fraud and Abuse Report, AI-powered bot traffic has surged 300% over the past year, while the commerce industry alone experienced more than 25 billion AI bot requests during a two-month period. As agent-generated traffic accelerates, the risk surface expands the overall need for verifiable identity becomes critical for merchants and customers to remain protected.
Trusted Agent Protocol addresses this need directly through the process of ensuring that every AI agent paying with a Visa credential is trusted, authenticated, and operating as intended. By utilising industry standard web infrastructure, Trusted Agent Protocol allows agents to transmit information to merchants to show that the agent is approved for its specific shopping mission, while also providing visibility into the consumer making the transaction, and securely passing payment information through a merchant’s preferred checkout flow. In addition, Trusted Agent Protocol was designed in order to scale with minimal infrastructure and user experience (UX) changes, making it easy for 175 million Visa accepting merchant locations around the world to adopt agentic commerce.
Akamai and Visa will also be working together to help merchants:
- Clearly identify a legitimate AI agent and its intent (meaning that Visa Trusted Agent Protocol is set to help merchants differentiate whether an agent is browsing or paying, while Akamai will strengthen that signal with real-time behavioral and network intelligence to detect anomalies)
- Confidently link the agent to the underlying user (the Trusted Agent Protocol will enable agents to pass pertinent information to connect each verified agent to the consumer it represents, and Akamai will preserve that identity through edge-based user recognition, maintaining the risk posture, trust signals, and account context merchants rely on in order to detect and prevent fraudulent activities).
- Enable secure, predictable payment interactions (the Trusted Agent Protocol is expected to assist agents with the process of conveying payment information in the manner a merchant expects an agent to pay, from network tokens to micropayments; at the same time, Akamai will reinforce these flows with end-to-end protection, as well as validating agent authenticity and helping to prevent fraud or abuse before it impacts the transaction).
