Agentic AI in Fraud and Risk Operations: What Actually Changes | Presented by AiPrise
Summary
AI is moving from buzzword to operational reality in fraud teams, automating identity verification, risk scoring, case review, and continuous monitoring, and even acting on decisions with limited human touch. This session separates genuine capability from hype, looking at where agentic AI meaningfully reduces manual review and false positives, and where it introduces new blind spots. It also tackles the harder question agentic commerce raises: how to verify and trust AI agents transacting on a customer's behalf. Attendees leave with a clear-eyed view of what to adopt now, what to watch, and how to keep control as automation scales.
Who should attend: Fraud prevention, risk, payments, and identity/onboarding professionals at merchants, marketplaces, PSPs, and acquirers — from analysts and managers running review operations to directors and VPs setting fraud and automation strategy.
What You'll Learn
- Distinguish real agentic capability from hype: what "agentic" actually means in fraud and onboarding workflows versus standard automation or rules.
- Identify where AI adds value vs. risk: which review, scoring, and monitoring tasks are safe to automate, and where human oversight still matters.
- Manage new failure modes: how to handle model drift, adversarial manipulation, and false positives that scale faster with automation.
- Verify identity and intent in agentic commerce: how to assess trust when an AI agent, not a person, initiates a transaction.
- Build an adoption roadmap: practical criteria for piloting agentic tooling without losing explainability or control.
Speaker
