Key Points
- Blockchain forensics firm Elliptic is using AI to improve detection of Bitcoin money laundering.
- The advanced tools have been tested on over 200 million transactions.
Elliptic, a firm specializing in blockchain forensics, has reported significant progress in employing artificial intelligence (AI) to detect money laundering involving Bitcoin.
The company is leveraging these advancements to bolster its detection tools. The details of this work have been outlined in a new research paper, jointly authored with scholars from the MIT-IBM Watson AI Lab.
AI in Crypto Forensics
The AI model used by Elliptic can successfully pinpoint proceeds from criminal activities deposited at a cryptocurrency exchange, identify new money laundering transaction patterns, and detect previously unknown illicit wallets.
Elliptic has put these enhanced tools to the test on a dataset comprising over 200 million transactions. The AI tools are trained to identify ‘subgraphs,’ which are chains of transactions that represent Bitcoin being laundered.
Blockchain and Machine Learning
This method allows for a more general focus on the ‘multi-hop’ laundering process, rather than the on-chain behavior of specific illicit actors. In contrast to traditional financial transactions, which are typically siloed, blockchains enable the more effective use of machine learning to scrutinize transactions.
Elliptic has a track record of identifying cryptocurrency transactions carried out by malicious actors, such as terrorist groups.
