Improving the quality of intelligence we can extract from blockchain technologies could vastly increase our capacity to counter financial crimes involving crypto or virtual assets.

This Q&A builds on a first article on the concept of blockchain intelligence and practical use cases. This one focuses on the quality of current blockchain intelligence and the need for recognised standards and professional training to build capabilities in the public and private sectors.

Thanks to Bogdan Vacusta (National Bank of Romania), Nico Di Gabriele (European Central Bank) and Victor Vevera (ICI Bucharest) for the insights. Note that the views expressed are personal and may not be necessarily aligned with those of the institution of affiliation.

Should we be worried about the quality of blockchain intelligence currently?

Yes. Since there are currently no established or empirically tested standards, there is a real risk that inaccurate blockchain “intelligence” is presented as fact.

To prevent errors and misattribution, we need rigorous standards to test the integrity of blockchain intelligence. Users must also understand its limits and its role in the overall analysis process.

How do investigators and analysts currently?

Many practitioners in the public and private sectors rely on commercial tools to trace transactions on the blockchain, but may lack understanding of how these operate or their limitations.

They may also simply rely on information that private blockchain tracing companies provide, with no means to independently verify the results and understand methodologies employed.

The information may be correct. But it’s a bit of a “black box” and the resulting blockchain intelligence may be of dubious quality – something that would never be accepted in traditional financial investigations.

Are you calling into question the quality of blockchain evidence used in court?

Yes. There are cases in the United States especially where the quality of the blockchain tracing methodologies has been questioned.

For example, in the case United States v. Sterlingov in 2024, the defendant was charged with operating the crypto mixer the Bitcoin Fog to launder illicit crypto assets and evidence in the form of blockchain analysis was provided by Chainalysis. The defence argued that “the [g]overnment's ‘blockchain analysis’ is junk science.” “The problem is we have no data set, no scientific data set with which we can measure the reliability and the accuracy of this software”, they said. This has triggered significant debate.

Where the quality of evidence from the blockchain is questionable, it may be dismissed – or even worse, result in unjust outcomes and potentially the conviction of innocent defendants.

Fortunately, in most cases so far, other evidence existed that enabled the prosecution to secure a conviction beyond a reasonable doubt. 

The quality of blockchain intelligence needs to be subjected to rigorous statistical evaluation and analysis. Where there is ambiguity, the burden to prove that adequate standards have been met should be on the party seeking to rely on the evidence.

How can law enforcement agencies build capabilities in blockchain intelligence?

It’s difficult!

• First, there are currently few options outside of commercial solutions. Training tends to focus on the specific solutions developed by the training provider.
• Second, when new staff are hired, especially in law enforcement and the public sector, these tend to be blockchain tracing experts who have gained their expertise using commercial tools. Again, this may lead to methodology bias.
• Third, the quality of available training varies significantly. Blockchain intelligence requires expertise in law, computer science and finance. Training courses rarely acknowledge the need to have at least a basic understanding of each of these fields.

Last, currently expertise is tied to certifications issued by providers of commercial tools, with no independent means to test the knowledge and skills of existing or new staff.

What can be done to improve blockchain intelligence tools, training and quality?

For tools, there is a pressing need for public-private partnerships to develop independently verifiable blockchain intelligence solutions. This would give all users more options. Working with universities, research organisations and other academic institutions is a good starting point to determine the reliability of blockchain intelligence and further research new applications.

For training, we need non-commercial blockchain intelligence training programmes adhering to the highest standards, with inputs and best practices from both public and private sector stakeholders.

For qualifications and hiring, we recommend that the role of a blockchain analyst is professionalised and officially recognised in frameworks such as the European Skills, Competences, Qualifications and Occupations (ESCO) classification. For that, an independent accreditation process for recognised professionals needs to be established, involving stakeholders from different sectors including independent accreditation bodies.

Underlying all this is the need to develop global standards based on a deep body of research that acknowledges both the capabilities and the limitations of blockchain intelligence.

A practical first step towards all of these goals is the creation of working groups of experts, academics and relevant organisations focused on the exchange and interoperability of blockchain intelligence, as well as on sharing technology and best practices.

Do initiatives to this end already exist?

Examples of such initiatives in which we are personally involved include:

• The Blockchain Intelligence Academy (BIA), whose mission is to develop “actionable, independent blockchain intelligence capability, in service of the truth by, placing “the integrity of blockchain intelligence above all else”. It evolved from a public-private partnership between ICI Bucharest and the Singapore-based blockchain intelligence firm ChainArgos. BIA is committed to developing training standards and is openly inviting all blockchain analytics providers to this initiative, as a key objective is to facilitate the use of various tools in training programs conducted by BIA.
• The Blockchain Intelligence Professionals Association (BIPA), which brings together practitioners in the field of blockchain technology. Its objective is to “establish a platform for dialogue, to facilitate training, sharing of expertise, good practice and to encourage new partnerships”. Bogdan Vacusta, the chairman of BIPA, is actively involved in facilitating interactions between blockchain analytics providers, with a key objective to facilitate public-private partnerships focused on standards and interoperability.

Both organisations are closely involved in the first ever Blockchain Intelligence Forum in Bucharest. Here, a special working group will be launched aimed at introducing a new occupation – Blockchain Analyst – into the Romanian and European classifications of occupations and qualifications.

The two initiatives are also exploring ways to provide blockchain intelligence accreditation that stakeholders globally respect and rely on.

Remind us: what is the dream? Our end goal?

The dream is to develop global standards of excellence for blockchain intelligence capabilities. Training is a top priority, to both strengthen the blockchain intelligence discipline and develop expertise that is less reliant on commercial blockchain tracing tools.

Our end goal is to establish a globally recognised standard for blockchain intelligence skills and develop the capabilities of professionals accordingly to combat the rise in illicit finance and preserve market integrity. A clean market is a precondition for its stability and thus capacity to fulfil people’s needs.

The rapid growth of crypto-assets and the development of blockchain technology has seen an increase in their use to facilitate financial crime. This is why it’s more important than ever to provide a structured, robust and resilient framework to continually improve and train blockchain intelligence professionals.