We would highlight this issue with help of a few notable recent (2025) Marketplace Data breach examples such as Zoomcar (The Indian peer-to-peer car-sharing marketplace disclosed a breach affecting 8.4 million users), Victoria’s Secret online store, and highlight just how vulnerable these platforms can be and where we should focus to address these vulnerabilities.

The next evolution in marketplace trust isn't just better detection, it's AI agents that reason about risk, negotiate remediation, and autonomously enforce with explainability. This session explores the shift from traditional ML pattern matching to agentic AI systems that understand context, predict seller behavior, and orchestrate evidence-based resolution pathways. Drawing from pioneering work integrating generative AI with enforcement at scale, you'll learn how platforms are building autonomous trust agents that reduce false positives through reasoning, handle appeal processes with human-like judgment, and satisfy emerging AI governance requirements like the EU AI Act, while maintaining the speed and scale modern marketplaces demand.

As artificial intelligence becomes more accessible and powerful, it is rapidly transforming the fraud landscape on both sides of the equation. The same technologies enabling businesses to scale faster and personalize experiences are also being leveraged by bad actors to create more sophisticated, harder-to-detect fraud—from deepfakes and synthetic identities to automated social engineering at scale. This panel brings together leaders from across marketplaces, fintech, and digital platforms to unpack this evolving duality. The discussion will explore how AI is being used to generate new attack vectors, how organizations are adapting their risk and trust frameworks in response, and what it takes to stay ahead in a world where fraudsters are increasingly AI-native. Attendees can expect practical insights on emerging fraud patterns, the role of AI in real-time detection and decisioning, and how companies are balancing growth, user experience, and security in an AI-driven ecosystem.

Traditional machine learning models are often slow to adapt to new fraud patterns and require large amounts of labeled data, while AI agents offer the potential for faster adaptation through few-shot learning and more human-like reasoning. This presentation explores a real-world production use case for reducing time-to-mitigation in fraud attacks by scaling post-transaction fraud investigations and enabling faster, more adaptive response mechanisms.

Take this time to visit the exhibitor space and connect with the solution providers and vendors on site. Explore the latest innovations in trust & safety, risk management, and legal compliance — all designed to help protect your platform, your users, and your business. Discover essential tools, build valuable connections, and stay ahead of emerging risks.