The Defense Information Systems Agency (DISA) released a request for information on artificial intelligence (AI) and machine learning (ML) technologies that can detect and combat cyberattacks as they occur.
The request, dated Oct. 23, explains that some of DISA’s current defense systems rely heavily on human intervention. The agency is interested in minimizing the time to “detect, respond to, and, ultimately, mitigate attacks” via an automated solution.
DISA seeks a program that becomes more sophisticated over time, limits impact on unaffected systems, detects false alarms, and categorizes the severity of each attack. Proposed solutions should also include “playbooks” defining ways to overcome known cyberattack techniques, but DISA expects to aid the service provider with adjusting these guidelines as adversarial tactics change.
In addition, the automation should be able to integrate with DISA’s existing tools, process terabytes of data each day, support over 1,500 organizations, and extract data from other systems.
By automatically responding to less complicated cyberattacks, the report asserts, an AI or ML-based cybersecurity procedure would free up valuable staff to focus on more complex issues. “DISA’s staff would be free to focus on harder and more immediate issues, leaving simple attack response to automated capabilities,” the request states.
DISA is accepting responses to the request until Nov. 8.