Blockchain analytics company Elliptic, in collaboration with the MIT-IBM Watson AI Lab, has released a research paper examining patterns of illicit activity within bitcoin transactions.
The study, utilizing a dataset known as Elliptic2, significantly larger than previous iterations, delves into the analysis of 200 million bitcoin transactions. This dataset expansion allows for a more comprehensive exploration of money laundering detection through machine learning techniques compared to their initial efforts in 2019, which were based on a dataset of only 200,000 transactions.
Within the Elliptic2 dataset, researchers identified 122,000 groups of connected nodes and transaction chains, referred to as “subgraphs,” associated with illicit activities. Through the application of artificial intelligence models, potential money laundering patterns within the Bitcoin blockchain were detected.
According to the findings, various techniques commonly employed in cryptocurrency money laundering were observed. Among these, “peeling chains,” where cryptocurrency is repeatedly sent to different addresses, and the use of “nested services,” which facilitate fund movement through cryptocurrency exchanges, were prevalent.
Tom Robinson, co-founder of Elliptic, highlighted the evolutionary nature of crypto laundering techniques, noting that the advantage of employing AI and deep learning approaches lies in their ability to automatically identify emerging patterns.
The research sheds light on the complex landscape of illicit activities within cryptocurrency transactions and underscores the importance of continued vigilance and technological advancement in combating financial crimes in the digital realm.
#Elliptic#MIT#IBM#Watson#AI#Blockchain#Research#Bitcoin#Transactions#SupplyChainNews