Intel, Georgia Tech to Lead DARPA Machine Learning Security Program

The Defense Advanced Research Projects Agency has tapped Intel and the Georgia Institute of Technology to oversee a four-year, multimillion-dollar program that will work to optimize cybersecurity defenses against potential attacks on machine learning and artificial intelligence models.

DARPA aims to develop defenses that will address potential scenarios that may affect an ML model to misinterpret data through the Guaranteeing Artificial Intelligence Robustness against Deception, or GARD, program, the company said Thursday.

The program is established to build theoretical ML system fundamentals that will characterize traits and validate gaps to reinforce a system.

“Intel and Georgia Tech are working together to advance the ecosystem’s collective understanding of and ability to mitigate against AI and ML vulnerabilities,” said Jason Martin, principal engineer at Intel Laboratories and principal investigator for the GARD program.