Los Alamos Scientists Develop AI to Fight Cryptojacking
Their neural network works faster and more reliably than non-AI systems, researchers said.
Aug 21, 2020 at 2:28 p.m. UTCUpdated Sep 14, 2021 at 9:46 a.m. UTC
:format(webp)/cloudfront-us-east-1.images.arcpublishing.com/coindesk/AFUTKDKLRBFC3P2TID4OXPEPWQ.jpg)
/arc-photo-coindesk/arc2-prod/public/LXF2COBSKBCNHNRE3WTK2BZ7GE.png)
Scientists at Los Alamos National Laboratory, the U.S. government-funded research outpost that once hosted the atomic bomb Manhattan Project, say they have designed an artificial intelligence for detecting would-be cryptojackers.
- In a press release, the scientists said their new AI sniffs out malicious code injections that can turn vulnerable supercomputers into zombie cryptocurrency mining operations, a serious IT issue that strikes governments and corporations globally.
- Called SiCaGCN, the neural network works by checking if a given program has the right backend structure to run on the computer system. Those that do, pass through. Those that don't, get flagged for removal.
- “This type of software watchdog will soon be crucial to prevent cryptocurrency miners from hacking into high-performance computing facilities and stealing precious computing resources,” project researcher Gopinath Chennupati said in the statement.
- SiCaGCN detected cryptojacking code faster and more reliably than non-AI solutions, according to the statement. The scientists originally proposed SiCaGCN in the journal IEEE Access last month.
:format(webp)/downloads.coindesk.com/arc/failsafe/user/1x1.png)