In The News

The three major security threats to AI


September 10, 2019

Less than one percent of AI research and development funding is going towards security, said Jason Matheny, Founding Director of CSET. He argues that more money needs to go into AI systems to protect them from three major types of attacks: adversarial examples, trojans and model inversion.

Accelerating threats to cybersecurity, the impact of automation on cyber defense, and the degree to which cyber operations will become faster and more powerful are among the subjects that CSET will now start to explore thanks to a grant from the William and Flora Hewlett Foundation.

CSET expert Ben Buchanan testifies to Congress on vulnerabilities and need for defense using the latest AI tools.

“These three areas—adversarial learning, cyber offense, and cyber defense—deserve a lot more attention, and quickly,” writes CSET Senior Faculty Fellow Ben Buchanan. “Policymakers have begun to consider these important issues, but the conversation must continue.”