A quantum computer that can crack today's state-of-the-art encryption algorithms is more than a decade off, according to the National Academies.
Gains in the accuracy of facial recognition technology over the last four years have been "revolutionary and not evolutionary," according to an expert from the National Institute of Standards and Technology.
In a world of real adversaries and malicious attackers, government can no longer live with employees' unforced errors.
Department of Homeland Security is calling on entrepreneurs and small businesses to help improve its emergency response, law enforcement, border protection and cyber risk management capabilities.
As emergency communications move to wireless broadband networks, the National Institute of Standards and Technology drafts a report that analyzes the cybersecurity of mobile and wearable devices.
Unlike in border control applications -- where the environment is controlled -- when supporting street policing, facial recognition algorithms make suggestions about possible similarities to police who must then confirm or refute them.
Leaked or stolen data is typically very public, but simple changes to a system are hard to detect even with a highly trained eye.
The general formula for cloud security is to determine the risks, mitigate them and then keep constant vigilance to spot and stop threats.
The Agency-Wide Adaptive Risk Enumeration algorithm is already ingesting data, and the Department of Homeland Security plans to put it into production heading in FY2020.
The Department of Homeland Security wants to prevent forgeries in the certificates and licenses issued by Customs and Border Protection, Citizen and Immigration Services and the Transportation Security Administration.
Researchers are developing algorithms that can detect when malware uses adversarial machine learning to attack networks and evade detection.
Defending against phishing attacks involves a combination of policies, procedures and controls.