Recently, data surpassed oil as the world's most valuable asset. Current data protection methods have too many dependencies on systems and networks through which data passes. So far, attempts to solve this problem have not adequately minimized external dependencies.
The self-protecting data concept, as a zero trust use case, involves adding protections to data objects to make such objects "self-protecting." The protections would include metadata tags and tamper-awareness and action logic that allows the data object to automatically, or remotely, choose courses of action when a given threat is detected. Artificial intelligence techniques are needed due to the complexity involved with managing numerous data attributes as metadata; the need for autonomous access control, infrastructure independence; and automation of detection, alerting, and response.
Describe basic requirements for a self-protecting data object.
Understand what research has been done so far on self-protecting data.
Understand how self-protecting data can leverage artificial intelligence techniques to improve data protection in zero or low-trust environments.
Operations Research Analyst, Data Analytics, Research & Innovation,
Headquarters Cyberspace Capabilities Center (USAF)