Blockchain
The elliptical shows how an AI model can identify Bitcoin laundering
This was stated by Elliptic, a blockchain analysis company that counts law enforcement, regulators and financial analysts among its clients uses a deep learning model, new AI techniques, and a massive dataset to better detect and monitor money laundering on a Bitcoin blockchain. The company is taking advantage of advances in graphical neural networks (GNN), which can process data that can be represented as graphs; GNNs have been used in cases such as drug discovery, computer vision, and natural language processing.
Because it is important? First, some context.
Cryptocurrency has been cited as a key enabler for ransomware groups. Allows the payment of ransoms Bitcoin, Ethereum or other virtual tokens that are difficult to trace. It can be hidden and recycled by means such as Crypto mixers. It can move easily across borders. And it allows bad actors to remain anonymous.
This concern goes back years. “Cryptocurrencies – which allow criminals to quickly extort large sums of money, can be anonymized and do not ensure consistent regulatory compliance, especially for attackers based abroad – have further enabled cyber criminals to commit disruptive ransomware attacks that threaten our national and economic security,” U.S. Senator Gary Peters (D-MI) said in a statement in 2022 after the release of a congressional report on the issue.
“You now have the ability to move millions of dollars in cryptocurrency across national borders in seconds,” Yonatan Striem-Amit, co-founder of cybersecurity provider Cybereason and its CTO at the time, told NPR in 2021.It really is a very powerful tool into the hands of criminals to launder money, to move currency from one state to another in a way that is somewhat untraceable and decidedly uncontrollable.”
Looking at the subgraphs
In a study conducted with the MIT-IBM Watson AI Lab, Elliptical researchers have focused on subgraph representationsa learning technique used to analyze local structures or shapes within a complex network, and applied to the analysis of illicit activity and money laundering on a blockchain.
“Rather than identifying transactions made by illicit actors, a machine learning model is trained to identify ‘subgraphs,’ chains of transactions that represent bitcoin laundering,” the company wrote in its blog post. “By identifying these subgraphs rather than illicit wallets, this approach allows us to focus on the ‘multi-hop’ money laundering process more generally rather than the on-chain behavior of specific illicit actors.”
Transparent transactions are key
While the use of cryptocurrency allows bad actors to remain anonymous, blockchains are transparent about their transactions and the types of entities conducting them, unlike traditional financial systems with siled transaction data. “While Bitcoin’s pseudonymity is an advantage for criminals, the public availability of the data is a critical advantage for those within law enforcement and financial institutions seeking to identify and investigate financial crime,” they wrote the researchers in the study.
The aim of the study was to show how companies and anti-money laundering investigators can use datasets that can identify the subgraphs they would be interested in, allowing them to separate the majority of subgraphs showing Bitcoin flows operated by legitimate services and those which include anomalous signs of money laundering-related activity.
New technique and a huge dataset
To help achieve this goal, Elliptic created a massive large graph dataset of nearly 200 million transactions. The dataset, dubbed Elliptic2, included 122,000 labeled subgraphs of Bitcoin clusters in a background graph of 49 million node clusters and 196 million edge transactions, the researchers wrote.
In contrast, five years ago, Elliptic – in a similar study involving the MIT-IBM Watson AI Lab and using a machine learning model to detect illicit Bitcoin transactions used by ransomware and other threat groups – used a dataset known as Elliptic1 of over 200,000 transactions.
The researchers teamed up with a cryptocurrency exchange to test the new technique and see if money laundering transactions could be identified. The technique identified 52 subgroups believed to be responsible for money laundering, only 14 of which had been reported by the exchange.
“Importantly, the exchanges’ insights were based on off-chain information, suggesting that the model can identify money laundering that would not be identifiable using traditional blockchain analytical techniques alone,” the company wrote.
Peeling chains and nested services
The AI model is based on a well-known money laundering model known as “peeling chains,” in which a crypto user sends – or “peels off” – a small amount of digital assets to one address and the rest to another address below user control. It also noted new patterns, such as the use of intermediary “nested services.”
“Nested services are businesses that move funds across accounts at larger cryptocurrency exchanges, sometimes without the knowledge or approval of the exchange,” the researchers wrote. “A nested service could receive a deposit from one of its customers to a cryptocurrency address and then forward the funds to its deposit address at an exchange.”
The model could also detect previously unknown illicit crypto wallets based on how the wallets’ funds were laundered, which they say could be used by law enforcement, financial regulators and blockchain analytics firms to more quickly identify such wallets .
Elliptic said it is making its dataset publicly available to help others create techniques to detect illicit crypto transactions.
Ransom payments increase
This will be important as the threat of ransomware and other financially motivated cybercrime grows. Blockchain analytics firm Chainalysis in a report this year called 2023 a “watershed year for ransomware”, pointing out that the amount was collected by bad actors ransom payments reached $1.1 billionsurpassing the previous record in 2021 by $983 million.
The FBI and other law enforcement agencies have successfully tracked stolen digital assets through the blockchain world, including last year, when the agency said it had achieved with encryption stolen by dangerous groups linked to North Korea. Affiliates linked to the TraderTraitor group have been responsible for the theft of hundreds of millions of dollars in cryptocurrencies from victims such as Alphapo, Atomic Wallet, CoinsPad and Harmony’s Horizon bridge.
The FBI did it too turn off several crypto mixers, which are services that blend illicitly obtained digital assets with other cryptocurrencies to obscure their origins. For example, stolen Bitcoins can enter a mixer alongside other tokens and come out as Ethereum, Monero, or combinations of cryptocurrencies.
Recent articles by author