AI-Accelerated Disinformation in Crisis Scenarios: National Security Implications and Federal Response

AI-accelerated disinformation poses a distinct and under-addressed threat to U.S. national security during periods of crisis. When a hurricane strikes, a conflict arises, or a public health emergency unfolds, the speed and scale at which AI-generated disinformation can spread directly undermines emergency response coordination, public compliance with safety directives, and the operational integrity of agencies tasked with crisis response. Additionally, while federal attention to disinformation has grown in recent years, policy efforts remain largely focused on disinformation’s impacts on elections and political processes (Alvisi et al. 2024), leaving a critical gap in how the government prepares for and responds to AI-accelerated disinformation during crisis scenarios. This memo identifies the mechanisms by which AI accelerates the spread of disinformation during crises, documents cases of measurable harm, identifies gaps in the current policy landscape, and proposes three federal interventions: (1) requiring DHS to incorporate AI disinformation threat assessments into national crisis preparedness planning; (2) directing the FCC and FEMA to establish a trusted source verification protocol and coordinate with major social media platforms; and (3) investing in AI-powered detection tools for state and local emergency managers through DARPA and DHS Science and Technology Directorate grants.