New report identifies types of cyberattacks that manipulate behavior of AI systems

Adversaries can deliberately confuse or even "poison" artificial intelligence (AI) systems to make them malfunction—and there's no foolproof defense that their developers can employ. Computer scientists from the National Institute of Standards and Technology (NIST) and their collaborators identify these and other vulnerabilities of AI and machine learning (ML) in a new publication.

ai and machine learning ai systems artificial artificial intelligence behavior computer cyberattacks defense developers identify institute intelligence machine machine learning nist report scientists security standards systems technology them types vulnerabilities