NEW YORK-- Riskified (NYSE: RSKD), a global leader in ecommerce fraud and risk intelligence, announced a major leap forward in its AI platform capabilities, introducing a powerful suite of control and empowerment tools designed to give merchant fraud teams complete visibility into risk patterns and identity behavior, conversational AI-driven insights, and the ability to surgically tailor their risk strategies. The innovation suite was announced onstage at Ascend 2026, Riskified’s premier global summit series, taking place May 4-6, 2026, at the Conrad New York Downtown in Manhattan, NY.

With global ecommerce continuing to expand and losses from AI-driven fraud attacks projected to spike, particularly with the rise of agentic commerce, where AI bots may conduct transactions on behalf of consumers, accurate fraud decisions are fundamental, but are no longer enough. Today, more than ever, sophisticated fraud and risk teams need to understand the why behind every transaction and pattern, and require the autonomy to instantly adapt to threats without risking their revenue.

“For over a decade, Riskified has been laser-focused on delivering the most accurate decisions through network data and deep identity models,” said Eido Gal, CEO & Cofounder of Riskified. “But our merchants told us they need more than just accurate decisions. They want to explore what's happening in their business in real-time, understand the context behind the data, gain insight into the identities behind it, and most importantly, take action with total confidence. We’re empowering merchants with unprecedented access to Riskified platform intelligence and the insights and ability to further tailor risk management to the needs of their business.”

Riskified unveiled the following advanced capabilities in its merchant control center that offer the most comprehensive, context-rich environment for fraud teams:

Riskified ARIA, the AI Risk Analyst

Going beyond basic chatbot add-ons and isolated investigation dashboards, Riskified ARIA (AI Risk Intelligence Analyst) gives merchants access to Riskified’s network data, model outputs, and identity signals, allowing them to better understand risk and fraud trends in their store at both a transactional level and across broader performance patterns. Riskified ARIA operates securely within Riskified's unified platform, tapping directly into billions of network signals, cross-platform patterns, and performance diagnostics. Fraud specialists can use simple conversational language to instantly zoom in on transaction-level explainability, visualize specific performance trends, or isolate specific risk indicators. ARIA serves as an always-on risk analyst that provides risk intelligence and insight across every touchpoint of the buyer journey in plain language and in seconds.

Identity Explore 2.0

Riskified is also expanding its identity capabilities beyond transaction-level analysis. Through the latest update to Identity Explore, fraud teams can move beyond individual accounts to a full view of identity across devices and across merchants in the Riskified network. It exposes hidden accounts and connections across the network and enables visibility into behavior patterns that cannot be seen within a single merchant view, giving merchants a network-wide view of identity and activity across multiple accounts and interactions, including suspicious behavior patterns, rather than operating in isolation. Fraud teams can see all their identities in one place, apply filters to expose cohorts of fraudsters and abuse rings, not just individual bad actors, and block or tag and track them in real time. Tapping into Riskified’s massive data consortium, this approach extends across the full customer journey, allowing merchants to assess risk based on how users behave, not just what they do in a single moment.

Decision Studio Enhancements

Decision Studio allows teams to turn what they learn from Riskified ARIA, decision explainability, and Identity Explore into deployable business rules. Intelligence-based fraud rules can be created, tested in simulation or shadow mode, and deployed after full confidence in the outcome. Business rules benefit from using identity and network patterns rather than single data points from a merchant’s own store, enabling a balance between a fraud team’s understanding of their business and Riskified’s AI and machine learning models. With the new “Identity Feedback Loop,” the identity and behavioral labels created in Identity Explore 2.0 become automated triggers for dynamic business policies in Decision Studio. Using these identity-based labels, merchants can refine their rules to better distinguish between legitimate customers and risky behavior flagged by a combination of identity signals with behavioral patterns. These enhancements put fraud teams at the steering wheel, allowing them to apply their unique business DNA to risk strategy with total confidence.

Together, these capabilities reflect a broader shift in fraud prevention from isolated decisions at checkout to continuous intelligence across the full customer journey. Rather than replacing automation, Riskified is focused on making it more usable, giving teams visibility into how decisions are made and practical ways to intervene when needed without slowing down operations.

These latest releases directly address the looming pressure from AI-driven fraud and abuse. Addressing visibility gaps created by siloed investigative tools, Riskified’s unified approach - spanning account intelligence, checkout intelligence, policy intelligence, and dispute intelligence - ensures that the most comprehensive and accurate insights are always embedded directly into merchant risk management workflows. This becomes even more critical as AI and agentic interactions grow, making it harder to distinguish between legitimate users and malicious activity without a clear understanding of underlying identity and behavior.

Riskified also shared new insights from a live survey conducted among hundreds of large enterprise ecommerce leaders at Ascend 2026, representing more than $1.1T in total processing volume (TPV). The live poll looked at how the merchants are progressing in their adoption of AI-powered shopping assistants as generative AI becomes more visible in the customer experience.

When asked about their current stage of adoption, responses showed a market still in transition:

· 12% already have a live AI shopping assistant
· 25% are actively building their own AI shopping assistant
· 41% are exploring their options but have not yet started building

Overall, 78% of merchants are already engaging with AI shopping assistants, whether they are exploring, building, or already live.

These results underscore a shift in ecommerce strategy as generative AI moves from backend optimization to front-end customer interaction, raising new questions around ownership of the customer relationship, experience control, and the role of external AI platforms in commerce.