On 8 October 2026 Riskified (NYSE: RSKD) announced Agent Identity Risk Intelligence, a set of capabilities the risk-and-fraud vendor says can identify the AI personal assistant behind an order or a customer-service request, resolve it to a real consumer identity, and return a risk signal in real time.
The problem Riskified is naming is delegated fraud: when a personal assistant such as Meta’s Muse checks out on someone’s behalf, the device and browser telemetry merchants normally rely on is largely absent. Where that telemetry is missing, Riskified says it leans on identity and network history, the behaviour of the agent (which signals intent), and — for post-checkout interactions such as refund requests — it can engage the assistant directly, confirming details or applying a “step-up friction that a legitimate assistant can satisfy and a script cannot”.
Rollout is staged. Riskified says the recognition will first be available to merchants on Shopify, where orders placed through Meta’s Muse already arrive tagged as agent-originated, and will extend to other platforms “as agent identification standards take hold”.
Why it matters for KYA
This is the KYA question arriving at a checkout till: not is this a valid login, but which agent is this, and who is accountable for it. As agent-originated traffic becomes a normal share of e-commerce, merchants need an answer to both halves — the agent’s identity and the human behind it — before they release goods or money.
Two things sharpen what is actually on offer:
- It is inference as much as identity. Much of the signal here is reconstructed from behaviour, network history and active challenge, not read from a credential the agent presents. That is a real and useful control, but it is a risk score about an actor, not a standardized, portable proof of who that actor is. Those are different jobs, and Riskified is candid that platform-level agent identification standards are still arriving.
- It is a vendor capability, not a framework. The value depends on coverage (Shopify first) and on how the underlying signals are combined. Treat the announcement as evidence that agent identity is becoming a line item in fraud tooling, not as a ratified way to verify an agent.
The durable lesson for practitioners is the accountability chain: when an agent acts, the useful record is agent → principal → scope, verifiable at the point of effect. Where that chain cannot be presented, merchants will keep rebuilding it by guesswork.
Sources
- Riskified, 8 October 2026: Riskified Launches Agent Identity Risk Intelligence: Know Who Is Behind AI Personal Assistants like Muse at Checkout and in Customer Service (company press release, ir.riskified.com).
- Business Wire, 8 October 2026: release distribution of the same announcement (businesswire.com).
- FF News / Merchant’s Eye, 8 October 2026: Riskified Launches Agent Identity Risk Intelligence to Combat AI-Driven ‘Delegated Fraud’ — trade coverage summarising the release.
Know Your Agent (KYA) explains agent identity, verification and accountability. This is an explainer, not legal or compliance advice — see our Sources & methodology. New to KYA? Start with What is Know Your Agent? and the glossary.