GAO Cites IoT, AI Among Long-Term National Security Threats

GAO found 26 long-range national security threats as identified by federal agencies, including the possibility adversaries could apply commercially available artificial intelligence to weapons. Other threats in Thursday's report include disrupting IoT-enabled critical infrastructure and devices; “developing autonomous capabilities that could recognize faces, understand gestures, and match voices of U.S. personnel, which could compromise U.S. operations”; and launching cyberattacks against critical and military infrastructure. Threat categories are: Adversaries’ Political and Military Advancements, Dual-Use Technologies, and Weapons, Events and Demographic Changes. DOD, the State and Homeland Security departments and Office of the Director of National Intelligence identified risks, and GAO reviewed national security documents and interviewed officials, it said. DOD told GAO the study provides “an accurate although sobering macro picture of how the US stands in the world against emerging threats.” The report is a public version of a classified one issued Sept. 28.