on the crypto
Los Alamos researchers say their AI system is faster and more reliable than non-AI-based methods.
Designed a new system to defend against illegal cryptocurrency mining
A group of researchers from Los Alamos, the US science lab that created the atomic bomb, has developed an artificial intelligence system that keeps crypto miners away from supercomputers.
From time to time, in fact, hackers manage to infect supercomputers, some of the most powerful computers in the world with processing speeds hundreds of thousands of times faster than even the best gaming PCs, using cryptocurrency mining machines and buy Bitcoins. The idea is to redirect all that power and use it to mine cryptocurrencies like Bitcoin.
System details published in a recent paper
This type of hack is the scourge of supercomputer operators, who preside over the computing power usually reserved for the best scientific research groups globally. In May, for example, at least a dozen European supercomputers had to shut down due to attacks from devices used for cryptocurrency mining, according to the BBC.
Those affected, or potentially at risk, may therefore appreciate the work of computer scientists at Los Alamos National Laboratory, who have devised an artificial intelligence system to stop such hackers.
Such a system is designed specifically for malicious actors who target supercomputers to mine cryptocurrencies such as bitcoin (BTC) and monero (XMR), the group specified.
How it works
The researchers explained it all in their article titled "Code Characterization With Graph Convolutions and Capsule Networks," which was published by IEEE, a leading Internet infrastructure body.
Unlike other systems that attempt to search for malicious code, the Los Alamos researchers' artificial intelligence system makes sure that the supercomputer only runs programs that should be running on its hardware.
In particular, the system is based on comparing graphs, which are like fingerprints for software. “Just as human criminals can be caught by comparing the spirals and arcs on their fingertips by recording them in a fingerprint database, the new AI system compares the contours in a program's flow control chart with a catalog of graphics for programs that can run on a given computer, ”the lab explained.
The researchers ran some tests and found that the system finds malicious crypto miners "much faster and more reliably than conventional non-AI based analytics."
