Riskified is an ecommerce fraud and risk company that helps online merchants approve more good customers while stopping fraud and abuse at every step of the shopping journey: login, checkout, refunds, returns and chargeback disputes.
Its machine learning models analyse each event in under a second and draw on identity signals gathered across a large network of merchant brands. The same signals now also help merchants judge orders placed by AI shopping agents.
At A Glance
| Vendor | Riskified |
|---|---|
| Category | Ecommerce fraud prevention and identity risk |
| Identities covered | Shopper identities and accounts, AI shopping agents |
| Agentic AI | AI Agent Intelligence, ARIA analyst |
| Customers | Lorna Jane, Rue Gilt Groupe, Gametime, Booking.com, Finish Line |
| Listing | New York Stock Exchange (RSKD) |
| Market classification | Identity Verification |
What Riskified Does
Online merchants lose money in two directions: to fraudsters, and to the good orders they decline because they cannot tell them apart. Riskified looks at account takeover at login, fraudulent orders at checkout, abuse of refunds, returns and promotions, and the chargeback disputes that follow.
The offerings share one identity view built from signals across the merchant network, so the same knowledge of a shopper informs account, checkout, policy and dispute decisions. Merchants also work with Riskified’s experts on a fraud and abuse strategy matched to their own business goals.
Products
Identity Risk Intelligence
Identity Risk Intelligence provides identity-level risk signals across the whole customer journey, giving the other offerings a shared picture of who is behind each session and order. Identity Explore 2.0 opens that picture to fraud teams as a network-wide view of identities across devices and merchants, surfacing hidden accounts and abuse rings that can be blocked or tagged in real time.
Account, Checkout And Dispute Intelligence
These three cover the transaction itself. Account Intelligence protects login and account activity, so merchants can stop account takeover without adding friction for genuine customers. Checkout Intelligence makes the fraud decision on each order in real time, and Dispute Intelligence supports merchants in fighting the chargebacks that follow.
Policy Intelligence
Retailers with generous refund, return and promotion policies use Policy Intelligence to detect customers who abuse them, while keeping those policies in place for everyone else.
Analyst Tools
Two tools help fraud teams act on these signals:
| Riskified ARIA (AI Risk Intelligence Analyst) | A conversational analyst that draws on network data, model outputs and identity signals to explain individual transactions, visualise performance trends and isolate specific risk indicators |
|---|---|
| Decision Studio | A rules environment for turning insights into business policies, testing them in simulation or shadow mode and deploying them; its Identity Feedback Loop uses identity labels to trigger dynamic policies |
How It Works
- Events in, decisions out. Login, checkout, refund and dispute events go to Riskified, and a decision comes back to the storefront.
- Policy layer. Merchant rules from Decision Studio sit on top of the model decision.
- Identity Graph access. Merchant AI assistants can query Riskified’s Identity Graph during a conversation through MCP, Google’s A2A protocol or standard REST APIs.
Identities And Systems Covered
| Shopper identities | Identity risk across login, checkout, refunds and returns |
|---|---|
| Customer accounts | Account activity, hidden accounts and abuse rings across devices and merchants |
| AI shopping agents | Orders placed by external AI agents such as ChatGPT, and merchants’ own native AI assistants |
| Transactions and disputes | Orders and chargebacks |
Agentic AI Capabilities
AI Agent Intelligence
AI Agent Intelligence covers orders from external AI agents and the native AI assistants merchants run on their own storefronts. AI Agent Identity Signals let an assistant check the Identity Graph for risk indicators mid-conversation, so it can decide instantly on a return or exchange request. The AI Agent Policy Builder, built on Decision Studio, lets businesses create, test, deploy and monitor rules for agent-placed orders, aimed at programmatic refund fraud, reseller arbitrage and promotion abuse. AI Agent Approve, an MCP server package, and Control Center dashboards let merchants approve and monitor orders placed by AI agents.
Company Background
| Founded | 2013, by Eido Gal and Assaf Feldman |
|---|---|
| Leadership | Eido Gal (Co-Founder and CEO), Aglika Dotcheva (Chief Financial Officer) |
| Offices | New York, Tel Aviv, London and Shanghai |
Note: This overview is compiled from Riskified’s publicly available information as of 4 November 2026. It is provided for information only. NHI Mgmt Group does not endorse any vendor, and nothing on this page represents the opinions of NHI Mgmt Group.