Researchers from the Department of Energy’s (DOE) Los Alamos National Laboratory (LANL) in New Mexico have developed an artificial intelligence algorithm intended to detect malicious codes that breach cryptocurrency data.
LANL said Thursday the algorithm is meant to identify codes that hack data from supercomputers to mine cryptocurrency forms such as Bitcoin and Monero.
The lab’s AI-driven approach utilizes graph comparisons and fingerprint-type identification to prevent hacker break-ins and seemingly legitimate programming that hide cyber threats.
Gopinath Chennupati, a researcher at LANL, said his team developed the AI algorithms in an effort to address recently reported cryptocurrency break-ins in Europe.
“Our deep learning artificial intelligence model is designed to detect the abusive use of supercomputers specifically for the purpose of cryptocurrency mining,” he said.
The LANL researchers’ findings are detailed in the “Code Characterization with Graph Convolutions and Capsule Networks” paper in the IEEE Access journal.