A new study, published by researchers from blockchain analysis firm Elliptic, MIT, and IBM, introduces a novel approach to identifying potential money laundering activities within Bitcoin’s blockchain. This approach utilizes an AI model trained on a dataset of 200 million transactions to detect patterns indicative of illicit money flows.
Rather than focusing on individual transactions, the researchers analyzed patterns of transactions leading from known entities engaged in illicit activities to cryptocurrency exchanges where they might cash out their funds. By identifying these patterns, the AI model aims to recognize the “shape” of suspected money laundering behavior on the blockchain.
The research team collected a substantial amount of data, comprising 122,000 patterns of known money laundering within the larger dataset of 200 million transactions. Using this data, they developed an AI model capable of identifying similar patterns across the entire blockchain.
In testing the AI tool, researchers found that it successfully identified suspicious transaction chains, with 14 out of 52 flagged accounts already under investigation by a cryptocurrency exchange for suspected illicit activity. Despite not having access to the exchange’s internal data, the AI model’s findings aligned with the exchange’s own assessments.
While the success rate may seem modest, the researchers argue that the AI tool significantly streamlines the process of identifying potential money laundering activities, reducing the number of accounts requiring investigation by a considerable margin.
Elliptic has already begun using the AI model internally, and the researchers have made the training data publicly available, fostering further research in the field. They emphasize the potential for this technology to enhance anti-money laundering efforts within the cryptocurrency space.
However, some experts caution that AI-based investigation tools may raise ethical and legal concerns, particularly regarding transparency and accountability. Despite this, proponents argue that these tools offer a more efficient means of detecting financial crimes and reducing false positives.
Overall, the researchers hope that their work will not only contribute to academic research but also provide practical tools for combating financial crime within the cryptocurrency ecosystem.
Stay current with supply chain report news at The Supply Chain Report. For international trade tools, see ADAMftd.com.
#CryptoAML#AI4ML#BlockchainDetection#CryptocurrencyAnalysis#MoneyLaunderingAI#SupplyChainNews
