When fraud stops looking like fraud: Why detection fails in cross-chain financial crime and what we must do next?

Financial crime is no longer easy to spot. In today’s blockchain ecosystems, particularly in cross-chain environments, illicit activity is increasingly designed to look legitimate. Rather than standing out, it blends in moving through networks in ways that appear consistent with normal user behaviour and, in doing so, bypassing many of the detection approaches we rely on today.

This session explores how and why this shift is happening. Drawing on real-world observations, it highlights how current techniques, including the use of cross-chain technologies such as bridges, are not about hiding activity, but about making it appear routine. As a result, detection methods that depend on identifying anomalies or outliers are becoming less effective against more sophisticated forms of financial crime.

The talk will examine where current approaches fall short, including rule-based monitoring, transaction analysis frameworks, and AI-driven models that rely on static patterns. It will also show how adversaries adapt quickly to detection mechanisms, creating an environment where defenders are constantly reacting to evolving behaviour.

More importantly, the session focuses on what needs to change. It introduces a different way of thinking about detection one that moves beyond "what looks suspicious" and instead considers how malicious behaviour can remain consistent with legitimate activity over time. This includes a stronger focus on behavioural context, visibility across systems, and the role of temporal and relational signals in improving threat intelligence.

Grounded in operational reality, the session will discuss what this means for security teams, financial institutions, and regulators. In line with the CyberCon theme of "Empowered Together," it highlights the need for closer collaboration across industry, academia, and policy to address increasingly complex adversarial behaviour.

Attendees will leave with a clearer understanding of why detection is struggling in cross-chain environments, how these gaps are being exploited, and what practical steps can be taken to strengthen detection in real-world settings.