NIST report: Mitigating the risks of cyberattacks on AI systems
By 
Adrianne Appel2024-01-11T21:11:00
Cyberattacks on artificial intelligence (AI) systems are increasing, so it’s important users know their vulnerabilities and try to soften the damage if they get hit, according to a new report by the National Institute of Standards and Technology (NIST).
There isn’t yet a foolproof method of preventing attacks on vulnerable AI systems, said computer scientists from NIST and other experts on adversarial machine learning.
Their 106-page report, published Jan. 4, describes common types of attacks, what types of vulnerabilities in AI systems are associated with them, and mitigations that can be employed when attacks occur. The report is intended to be paired with NIST’s AI Risk Management Framework, a document that provides guidance to AI users and developers about what voluntary steps regulators want to see in terms of cybersecurity of AI systems.
THIS IS MEMBERS-ONLY CONTENT
You are not logged in and do not have access to members-only content.
If you are already a registered user or a member, SIGN IN now.
Related articles
-
News BriefNYDFS proposes AI use guidance for insurers
The New York State Department of Financial Services issued for public comment guidance for insurers operating in the state regarding their use of artificial intelligence systems and other predictive technologies.
-
PremiumAI in 2024: More business use, more fraud risks
Use of generative artificial intelligence by businesses will ramp up in 2024, as will risk of AI-driven cyberattacks and fraud, according to experts.
-
PremiumBiden AI executive order tips hand on areas of regulation focus
It’s all hands on deck at the White House to put into motion the dozens of directives in President Joe Biden’s executive order on artificial intelligence, according to Nik Marda, chief of staff for the Technology Division in the Office of Science and Technology Policy.
More from Cybersecurity
-
News BriefU.K. says company boards need to worry more about cybersecurity risks
The U.K. government wants directors and boards of directors to become more actively involved in cybersecurity risks facing public and private companies, as the world faces “alarming” threats from criminal gangs and malicious nation-states. Though many organizations take cybersecurity seriously, the U.K. government says they do not place management of ...
-
PremiumNavigating compliance: A guide for small teams to tackle CMMC
Many small organizations within the Defense Industrial Base are struggling to meet the rigorous requirements validated through the Cybersecurity Maturity Model Certification, writes Thomas Graham, CISO at Redspin. If you haven’t been tracking it closely, CMMC was finalized in October, with an effective date of December 16, 2024.
-
PremiumFinancial crime in the shadows of the dark web
The dark web has been depicted as a long-standing hub for crimes, where illegal activities such as drug dealing, financial fraud, weapon sales, murder for hire, stolen credit cards, and ransomware gags are easily accessible to the public.