Elliptic Taps Artificial Intelligence (AI) to Crack Down on Bitcoin Money Laundering

Blockchain analytics firm Elliptic said its researchers have made progress in leveraging artificial intelligence (AI) to detect money laundering in Bitcoin.

A recent paper detailing this research was co-authored by researchers from the MIT-IBM Watson AI Lab. Elliptic has revealed the successful identification of illicit proceeds deposited on a cryptocurrency exchange, new patterns of money laundering transactions, and previously unidentified illicit wallets using a deep learning model.

The company said the results are already being used to improve its products.

Exposing hidden money laundering patterns

The underlying data has been made public. With more than 200 million transactions, this dataset will allow the community to create new ways for artificial intelligence to identify illegal cryptocurrency activities, Elliptic said in a statement. statement this week.

Instead of identifying transactions being made by criminals, the machine learning model is trained to recognize “subgraphs” – essentially strings of transactions that indicate Bitcoin laundering. This method focuses on identifying these subgraphs rather than illicit wallets, allowing Elliptic to focus on the broader “multi-hop” money laundering process rather than the specific on-chain actions of individual criminals.

Elliptic said it tapped an undisclosed cryptocurrency exchange to test whether the technology could detect money laundering attempts through it. Of the 52 predicted “money laundering” subgraphs ending in deposits on this exchange, 14 have been confirmed by the exchange as being associated with the flagged users. On average, less than one in every 10,000 accounts are flagged, indicating strong performance of the model.

“We have barely scratched the surface of what is possible in this area, but this work has already led to benefits for Elliptic users. Further collaboration and data sharing will be key to developing these technologies further and combating financial crime in crypto assets.”

The role of artificial intelligence in the blockchain sector

AI tools are slowly showing an extraordinary ability to analyze vast data sets to identify patterns beyond human understanding, such as identifying illegal financial movements within the Bitcoin economy.

Therefore, it is not surprising that venture capital investment in Web3 and AI startups will exceed $637 million in 2023.

In fact, AI agents are expected to dominate the blockchain sector in 2024 to create a more secure and efficient ecosystem, as stated in a report by Nansen.