## A 1LoD report ### FORECAST 2026 # One Financial Crime Risk with Many Channels --- Key Takeaways Firms have entered the crypto and digital asset space cautiously, using phased, partner-led models to manage regulatory uncertainty rather than launching full-scale propositions. Anticipated regulatory shifts, including a more permissive US posture, act as a catalyst for experimentation but not a relaxation of anti-money laundering (AML) standards. Governance models evolve by extending existing AML, sanctions, and fraud frameworks to digital assets, rather than creating entirely new paradigms. The most significant structural risk remains the identity gap between on-chain activity and real-world actors, particularly in decentralized finance (DeFi) and off-platform environments. Operational effectiveness depends less on acquiring blockchain analytics tools and more on integrating, enriching, and operationalizing data for investigators. Risk ownership and accountability across the 1st and 2nd lines remain a point of tension as digital asset activity expands. Crypto-native firms demonstrate materially greater speed in deploying models and controls, but that agility requires compensating governance, QA, and explainability. Regulatory fragmentation and uneven supervisory capacity continue to enable arbitrage and constrain the scalability of global financial crime programmes. --- # One Financial Crime Risk with Many Channels **The Financial Crime Leaders’ Network brought together senior leaders in financial crime in New York on 09 October. Firstly, participants focused on the extent to which financial institutions had begun to de-silo financial crime functions and the practical realities of delivering a more holistic approach to financial crime risk management. While there was broad agreement that integration across anti-money laundering (AML), know your customer (KYC), fraud and sanctions was both desirable and inevitable, participants consistently emphasised that progress had been incremental, uneven and far more advanced at the data layer than at the organisational level.** Many firms had moved towards integrated investigative teams across AML and sanctions, often supported by shared case management tools, but fraud typically remained anchored in the 1st line with its own investment priorities and technology stack. As one participant observed, “you can work cases together regardless of the organisational structure, but if the data doesn’t flow, you’re still operating in silos.” Examining the Limits of De-Siloing Financial Crime Functions A recurring theme was that de-siloing was less about collapsing reporting lines and more about enabling a common view of risk through shared data and analytics. Several participants described efforts to build single data domains or global data lakes to bring together transaction, customer and behavioural data across business lines. However, this was widely characterised as a multi-stage journey, beginning with basic data harmonisation, followed by the development of cross-functional investigative processes, and only later by the removal of redundant systems and duplicated monitoring. Participants stressed that without getting the foundational data layer right, firms would “never get to the point where you can actually turn systems off or stop investigating the same transaction three times in three different teams.” Data integration challenges dominated the discussion. Banks reported persistent fragmentation between KYC repositories, transaction monitoring (TM) platforms, fraud systems and sanctions screening tools, often built for different purposes, at different times and under different regulatory pressures. Investigators frequently lacked access to core KYC information, while contextual intelligence generated by financial intelligence units (FIUs) was not systematically fed back into onboarding or ongoing due diligence processes. Funding and prioritisation emerged as critical constraints and participants were candid that, in many institutions, large-scale integration only became feasible when driven by regulatory enforcement. Several cited consent orders as the trigger that unlocked investment in enterprise data platforms and enabled consistent data feeds across 1st and 2nd line functions. Without such pressure, integration initiatives were frequently deprioritised in favour of mandatory remediation work. The role of financial crime compliance leadership was a central point of debate. Participants generally agreed that compliance leaders were expected to champion integration, articulate the long-term risk management case and challenge siloed thinking. However, there was also recognition that compliance had, in some cases, taken on too much responsibility for decisions that should have sat with the business. Several participants warned that when technology and financial crime teams defined authoritative data sources and system hierarchies without sustained business ownership, accountability became blurred. Widespread discomfort around data ownership, with the 1st line often reluctant to accept accountability for customer data beyond what was strictly required to transact. Some attributed this to incentives, while others framed it as a cultural issue shaped by past enforcement patterns. There was consensus that future governance models would need to place clearer accountability on the 1st line for data quality and outcomes, with compliance setting standards, oversight and challenge rather than acting as de facto data owner. Governance risks were seen as intensifying as firms adopted more advanced technologies, particularly AI and machine learning (ML). While participants were cautious about vendor claims and stressed that much of what was currently deployed was rules-based automation rather than true AI, they acknowledged that new tools were increasing the distance between customer-facing teams and underlying decision-making logic. Several warned that without clear governance over data lineage, decision rights and change control, firms risked future client friction and regulatory scrutiny. Overall, the debate painted a picture of an industry aligned on the direction of travel but constrained by legacy, incentives and governance complexity. De-siloing was progressing, but primarily through data-sharing initiatives rather than wholesale structural change. Participants agreed that achieving a genuinely holistic financial crime risk framework would require sustained leadership, clearer ownership models and governance structures capable of supporting shared platforms over the long term, rather than relying on regulatory pressure alone to force integration. --- # Balancing Speed, Innovation and Accountability as Digital Assets enter Regulated Financial Institutions **The second debate explored how financial institutions and crypto-native firms had been preparing for a shifting regulatory landscape and what that had meant for crypto-related AML governance.** Several firms described adopting deliberately constrained business models as a form of regulatory and operational risk management. Rather than launching broad wallet functionality, some banks had partnered with exchanges so that customers routed orders externally while funding accounts through existing brokerage relationships. This gradual approach had also reflected concerns about supervisory perception. The discussion then turned to how governance models had evolved traditional AML and fraud frameworks to address blockchain-based risks and DeFi-adjacent activity. One participant argued that they had built a compliance programme that “looked and felt like what you have at banks,” combining transaction monitoring, sanctions controls and KYC, albeit delivered differently for competitive and operational reasons. They explained that crypto controls still needed to be auditable and defensible. That ability to move quickly was presented as a defining contrast with traditional banking governance. However, speed had not been portrayed as inherently safe; the exchange described having to persuade the business that rapid delivery still required governance and controls embedded into the development process. Quality assurance and quality control functions had been strengthened to test both what machine learning surfaced and what it suppressed. On risk ownership and expertise, participants repeatedly implied that digital asset capability had been spreading across both 1st and 2nd lines, but the practical challenge had been timing, resourcing, and clarity of accountability. Some practitioners had felt structurally disadvantaged when product teams moved too quickly. When the debate addressed detection and investigation challenges, the group converged on a consistent operational reality: the blockchain provided excellent transactional visibility, but attribution remained difficult without identity data and information-sharing. Firms described using a common set of blockchain analytics vendors alongside homegrown attribution capabilities to link addresses to individuals or entities. The discussion highlighted a major ecosystem weakness: competitive dynamics had inhibited cross-exchange cooperation, meaning there was no consistently effective analogue to established information-sharing norms in banking. This lack of sharing, combined with fragmented regulation across jurisdictions, was seen as a key reason why illicit actors could exploit gaps and pursue the “path of least resistance.” Finally, participants emphasised that the biggest operational challenge was making blockchain data usable at scale. This work was described as “not exactly rocket science, it’s just very tedious” largely because building and maintaining the necessary APIs and integrations was complex. Without strong user experience, even rich datasets could produce slow investigations, creating a programme that had data “but you can’t actually drive” effectively. This information was taken from The Financial Crime Leaders' Network event in New York, 9 October 2025.