Written by: Will Awang
The current field of anti-money laundering (AML) for cryptocurrencies is primarily dominated by blockchain analytics tools (BATs), such as Chainalysis and Elliptic. Although these tools have strong on-chain capabilities, once cryptocurrencies are more integrated with real-world scenarios—stablecoin payments—how can we ensure compliance with anti-money laundering regulations while safeguarding on-chain funds/wallets? This is a key concern for expanding real-world scenarios for stablecoin payments and is a question that must be answered for AML compliance.
Although cryptocurrency exchanges can currently use blockchain analytics tools (BATs) and bind the strong KYC information of users on their platforms—KYA wallets + KYT assets + KYC user information—to achieve relatively compliant anti-money laundering measures and risk monitoring, there are still vulnerabilities for C-end consumption scenarios of stablecoin payments, especially in the current social entertainment scenarios of stablecoin retail payments.
A recent study by Singapore licensed digital asset service provider MetaComp has exposed the 'fig leaf' of industry risk control. The report tracked over 7,000 real on-chain transactions and found that relying on only 1–2 KYT (Know Your Transaction) tools for screening would lead to approximately 25% of high-risk transactions being misclassified as 'safe' and allowed to proceed. In other words, a quarter of potential threats slip through unnoticed—this is no longer a 'blind spot' in risk control but rather a 'black hole' devouring risk.
Therefore, on-chain anti-money laundering compliance cannot merely operate within the cognitive blind spots of using a single blockchain analytics tool (BATs): looking at the world with one eye also requires combining it with off-chain fiat anti-money laundering (AML) transaction monitoring solutions. Thus, we compiled the article 'Deloitte: Conquering Crypto Crime' to explore a solution that combines blockchain analysis with real-time transaction monitoring, which we believe will further drive the implementation of stablecoin payment scenarios.
Introduction
Blockchain technology and cryptocurrencies have gained significant attention in recent years. However, the pseudo-anonymity—sometimes even complete anonymity—of these technologies makes them susceptible to misuse for money laundering activities. Blockchain analytics tools (BATs) are currently at the forefront of combating money laundering threats in the cryptocurrency industry. Companies such as Chainalysis, Elliptic, and Scorechain collect risk data regarding pseudo-anonymous and anonymous blockchain addresses and conduct clustering analysis to consolidate insights.
For instance, if a blockchain analytics tool (BAT) identifies a blockchain address associated with a ransomware attack, it will absorb and store that address along with its associated risk information. However, despite the effectiveness of these tools, current methods and tools still have several limitations:
End-to-end monitoring of fiat and cryptocurrency assets: BATs often lack the ability to trace the flow of exchanges between fiat and cryptocurrency, which is a key link in the money laundering chain.
Pattern analysis: BATs typically can only identify suspicious transactions or behavior patterns based on simple rules, making it difficult to uncover more complex anomalies, which does not meet regulatory requirements and limits the depth of money laundering identification.
Indirect risk scoring: Since only a few blockchain addresses are directly associated with confirmed criminal activities, BATs often provide only indirect risk scores. For these indirectly scored addresses, analysts often face significant subjective judgment when deciding whether to submit a Suspicious Activity Report (SAR).
In addition to a low detection rate (thereby increasing the risk of being used for money laundering), these limitations may also lead to technical deficiencies in complying with current anti-money laundering laws and regulations. Regulatory authorities often require adherence to known typologies issued by financial intelligence units (FIUs) and/or law enforcement agencies, either directly or indirectly. The typologies published by financial intelligence agencies and the Financial Action Task Force (FATF) often cover mixed transactions of fiat and cryptocurrencies as well as more complex behavior patterns, scenarios that cannot be fully covered by blockchain analytics tools (BATs).
Additionally, the aforementioned shortcomings can lead to inefficient alert processing. Addresses that have indirect associations with tainted addresses and are in a medium risk range often trigger a large number of alerts but provide limited additional risk information. These alerts are usually classified as false positives, forcing operational personnel to spend considerable time dealing with ambiguous alerts—efforts that often turn out to be futile.
1. Fill the gaps of blockchain analysis with real-time transaction monitoring
The aforementioned limitations highlight the necessity of complementing blockchain analytics tools (BATs) with anti-money laundering (AML) transaction monitoring systems that are widely applied in the fiat sector. These systems are designed specifically to handle fiat transactions and can analyze based on complex scenario models. Combining BAT with AML transaction monitoring systems 'adapted for cryptocurrency assets' can jointly overcome their respective shortcomings.
1.1 Connecting Fiat and Cryptocurrency Transactions
To achieve effective risk management, monitoring systems should associate fiat and cryptocurrency transactions to present the overall behavior of customers, for example, by integrating via customer numbers or account IDs. Particularly in cryptocurrency exchanges that have both types of transaction data, this approach can reveal insights that cannot be discovered through a single dimension (only cryptocurrency or only fiat). Key typologies focus on the complete cryptocurrency customer journey: from depositing fiat currency, trading various cryptocurrencies, to ultimately withdrawing fiat to another bank account.
1.2 Analysis of Complex Transaction Patterns
To achieve efficient and effective money laundering detection, it is necessary to identify known (but undetected) complex money laundering behavior patterns. Although modern BATs allow for rule creation, they are often limited to simple and usually one-dimensional criteria (such as amount thresholds). More complex rule sets, or even AI-based models—as adopted by many modern AML transaction monitoring systems—can help operational personnel discover common and complex suspicious behaviors such as 'Money Muling' and 'Account Passthrough.'
1.3 Incorporating Customer Risk Data
Although the traditional fiat sector has not yet fully integrated static 'Know Your Customer' (KYC) data with dynamic transaction data, modern AML transaction monitoring systems support such an integrated view, allowing operational personnel to utilize all available customer data for optimal risk detection. In the cryptocurrency realm, due to the pseudo-anonymity of blockchain transactions, the correlation between customer data and transaction data is particularly crucial. To achieve sufficient risk detection and cover all typologies released by regulators, it is necessary to integrate static customer data, dynamic transaction data, and risk information provided by BAT.
2. What kind of compliance system do we need?
Blockchain analytics tools (BATs) serve as a foundation for reducing money laundering risks associated with cryptocurrencies, but they also need the supplementary functions provided by modern transaction monitoring systems. Integrating cryptocurrency market tools with modern monitoring systems is the future roadmap for achieving efficient and effective AML compliance. Using AI-based models to identify known and unknown patterns and reorder the preliminary risks identified by BAT is not only meaningful but may also be a trend. We advocate for the coordinated and comprehensive use of BATs and modern traditional fiat monitoring tools to achieve efficient and effective anti-money laundering compliance, ultimately curbing the misuse of blockchain technology and cryptocurrency transactions.
Specifically, we recommend the following ways to integrate BAT with modern financial crime technology systems:
2.1 Unified User Interface
An integrated user interface allows operational personnel to seamlessly access and present all information required for analysis and verification of transactions on a single, synchronized platform, rather than switching back and forth between multiple standalone applications. Centralizing customer data, fiat and cryptocurrency transactions, complex behavior patterns, and risk prioritization models can significantly enhance processing efficiency.
2.2 Joint Presentation of Fiat and Cryptocurrency Cash Flows
This core recommendation stems from the value of presenting both types of transactions in a unified format. Criminals often attempt to reinject illegal gains back into the legitimate financial system; therefore, effective monitoring must associate and analyze fiat and cryptocurrency transactions together. When operational personnel can view this complete linkage, they can better identify and mitigate complex money laundering patterns.
2.3 AI Models for Detection and Result Ranking
AI models can enhance efficiency and accuracy through anomaly detection and risk analysis-based alert prioritization. Additionally, these models can dynamically reorder the preliminary findings of BAT based on multi-dimensional risk indicators; in cryptocurrency transactions, funds often go through multiple consecutive transfers, and risk scores often linger in the 'grey area,' making AI ranking particularly important. It is worth emphasizing that the models used should be transparent and interpretable; especially for regulatory purposes, AI-generated decisions must be accompanied by human-readable explanations.
Complex trading patterns and customer risk data can hinder the effectiveness of cryptocurrency-related AML compliance. However, by integrating the user interface, jointly presenting fiat and cryptocurrency cash flows, and using AI models for result ranking, the efficiency and quality of monitoring processes can be significantly improved. Therefore, integrating blockchain analytics systems with traditional AML systems is the best way to enhance the efficiency and effectiveness of anti-money laundering compliance in the cryptocurrency space.
3. Deloitte Case Study—From Independent Assessment to Integration Support
Before remodeling any transaction monitoring system, we first conducted an independent assessment of transaction monitoring in both crypto and non-crypto customers. The scope of the assessment can be flexibly adjusted by module, covering different regulatory focus areas such as analyzing underlying policies and processes, data handling, coverage, and data lineage.
In this case, we focused on examining the coverage of associated risks and typologies, and assessing the effectiveness and efficiency of existing rules and parameters in detecting these risks. In addition to analyzing internal documents (such as risk assessment reports), we also specifically measured the extent to which publicly available cryptocurrency-related typologies (from various financial intelligence agencies and FATF) are implemented across the entire transaction scenario.
Based on a priority list that includes over 50 applicable typology scenarios, we found that some scenarios are completely uncovered, while most are covered but rely on highly manual and decentralized methods. Almost all key information (crypto and fiat transactions, other customer data) already exists but is scattered across different systems, making it nearly impossible for alert processors to 'connect the dots.' Further analysis of the alert clearing processes and alert statistics shows that operational personnel spend most of their time sifting through false positives; meanwhile, the significant suspicious behaviors identified proactively by transaction monitoring systems are almost zero. Combined with feedback from law enforcement and other regulatory requests, we know that there are indeed suspicious behaviors within our known customer base, further confirming the urgent need to adjust detection logic and alert prioritization.
After completing an independent assessment, we assisted the client in developing a rectification plan: first mapping out the Target Operating Model, based on which we evaluated the feasible options of 'in-house' or 'outsourced.' During the assessment phase, we helped the client dissect and refine requirements based on all dimensions within the assessment framework.
Ultimately, the customer independently completed the vendor selection. Subsequently, we intervened in the integration project as needed in various roles: the primary task was to break down the relevant fiat-cryptocurrency typologies into actionable rules and model configurations, ensuring their effectiveness and efficiency through testing; secondly, we provided support from both business and technical fronts: on the business side, assisting in mapping existing interfaces to the new vendor's interface; on the technical side, assisting in API configuration and constructing necessary connectors.
Ultimately, we helped clients achieve:
Be based on facts, clearly define (regulatory) issues;
Select the optimal solution in the market;
Complete integration within a minimized time window, without overburdening the client's already strained resources.
4. Hawk Case: Integrated Monitoring of Fiat & Cryptocurrency
Hawk has developed a cryptocurrency asset monitoring solution that deeply integrates its own AML transaction monitoring technology with the capabilities of blockchain analytics tools (BATs), and provides modules for customer risk rating, customer screening, payment screening, and more.
4.1 Connecting Fiat and Cryptocurrency
The Hawk platform can read various transaction data—covering standard fiat transaction protocols such as SWIFT, SEPA, as well as cryptocurrency transaction data. The system can simultaneously ingest fiat and cryptocurrency transactions, allowing for cross-data source identification of overall patterns.
For example: A new customer at a cryptocurrency exchange deposits a large amount of fiat currency, quickly exchanges it for various cryptocurrencies, and then transfers small amounts of the split crypto assets to multiple external addresses within a short period. Since BATs can only track on-chain cryptocurrency flow and cannot capture this classic 'layering' technique—often used by money launderers in exchanges—Hawk can comprehensively identify the entire money laundering chain by synchronously monitoring fiat and cryptocurrency transactions.
4.2 Merging Blockchain Analysis with KYC Customer Risk Information
In addition to processing fiat and cryptocurrency transactions simultaneously, Hawk has also established standardized integration with BAT providers, embedding BAT risk scores into its rules and models to provide additional insights for anti-money laundering.
Taking the aforementioned case as an example, applying the BAT wallet risk score to known money laundering schemes can provide further clues for external wallets used to extract funds. These wallets may be directly or indirectly associated with addresses that have been marked for illegal activity, such as the receiving address for ransomware proceeds. As a result, alerts will gain richer contextual information, helping operational personnel to make more accurate judgments about whether the behavior is suspicious.
The out-of-the-box Hawk solution will also automatically inject other relevant customer risk information (KYC) into alerts. When the system detects suspicious activities, it will synchronize additional background information (such as geographic location, occupation information, watchlist, and negative media search results). For example, if a client registers in Germany but logs in from a device using an IP address from Vietnam, these details will provide valuable context for the investigation.
Ultimately, all information is consolidated into a unified case management system, presenting transaction and customer-related information in a complete and one-time manner. This system also supports maximum automation of subsequent processes, such as automatically generating and submitting Suspicious Activity Reports (SARs).
4.3 Overlay AI, entering a new era of cryptocurrency transaction monitoring
A core task of Hawk is to overlay transparent and interpretable AI on all relevant data to further enhance the effectiveness of transaction monitoring. Hawk's AI model specifically incorporates cryptocurrency signals into detection and false positive reduction models, fully leveraging transaction data, customer information, and risk data provided by BAT.
This practice not only allows models to accurately identify illegal money laundering schemes but also addresses the core challenge in current cryptocurrency transaction monitoring: many results provided by BATs fall into a 'grey area,' making it difficult for analysts to determine if they are truly suspicious. Hawk's cryptocurrency model integrates the latest data analysis technologies and utilizes all transaction and client data, enabling operational personnel to accurately filter these 'grey' alerts and avoid wasting significant manpower.
5. Conclusion
We discussed the limitations faced when relying solely on blockchain analytics tools (BATs) like Chainalysis, Scorechain, or Elliptic for anti-money laundering (AML) efforts. These tools, while powerful, are still insufficient for comprehensive money laundering detection: they cannot handle fiat transactions and struggle to identify suspicious transactions or behavior patterns; their risk scoring methods are mostly indirect, leading to inefficient alert clearing processes.
To break through the bottleneck of BATs, we recommend supplementing them with traditional AML transaction monitoring systems that can link fiat and cryptocurrency transactions and support complex transaction pattern analysis. At the same time, customer risk data should be taken into account. Specifically, we advocate for the deep integration of BATs with modern financial crime technology systems: Seamlessly presenting and accessing transaction analysis data through a unified user interface; presenting fiat and cryptocurrency cash flows in a single format; and using AI models to detect anomalies and intelligently rank results.
We also demonstrated viable pathways through case studies: Deloitte assisted clients in comprehensively assessing their transaction monitoring systems; Hawk merged its self-developed AI enhancement solutions with BAT capabilities, significantly improving monitoring effectiveness.
In summary, we advocate for the coordinated use of BATs alongside traditional fiat monitoring tools to achieve efficient and effective anti-money laundering compliance in the cryptocurrency space.
