DISA Aims to Use Artificial Intelligence, Machine Learning for Cyber Defense

The Defense Information Systems Agency is pursuing efforts aimed at utilizing artificial intelligence and machine learning to boost cyber defense.

One of those efforts is underscored in a cooperative research and development agreement with NT Concepts.

According to DISA’s Roy Hendrix, work under the agreement will be focused on identifying candidate state-of-the-industry ML technologies and figuring out how they can be used to deliver operational insights and strengthen cyber defense mechanisms across the Department of Defense Information Network.

Hendrix said the goal is to “leverage AI/ML computing capabilities to automate data pipelines and use them for both performance and cybersecurity,” Signal Magazine reported.

DISA also started a pilot program on applying AI/ML capabilities to detect cyber anomalies and adverse activity, and provide incident response guidance.

The agency will conduct an initial exploratory data analysis to see where to apply the algorithms, which, according to the pilot’s technical lead, Deepak Seth, will take at least six to nine months.

Seth told Signal Magazine that DISA will look at whether AI models or algorithms trained on historic agency data can generate predictions that can help analysts in incident response and investigations.

The Joint Artificial Intelligence Center’s Joint Common Foundation, a development environment for testing, validating and fielding AI capabilities, will play a key role in the pilot.

Seth noted that JCF will enable DISA to develop an AI capability and apply it to data being collected from big data platforms.