AI Is Quietly Creeping Into Emergency Alerts, and Experts Aren't Sure the Public Is Ready

Emergency management experts say generative AI is likely already widespread inside state and local disaster agencies, even where official policy forbids it. The warning comes after a county emergency manager in Alabama used an AI chatbot to produce a graphic about where Hurricane Isaias might threaten the coast with a "sting jet" of high winds. According to Laura Myers, director of the University of Alabama's Center for Advanced Public Safety, the graphic was not meteorologically accurate.

Myers and others say the appeal is understandable. Emergency managers are often short on staff and resources, and many write alerts full of jargon, assume the public already knows key details, or bury the important information. Jeannette Sutton, a disaster researcher at the University of Albany, said agencies recognize public communication as a priority but don't act on it until they realize they should have.

Bad warnings can cost lives. In Montecito, California, after the 2017 Thomas Fire, a rainstorm triggered a mudslide that killed 23 people and injured more than 160. Sutton said governments that issue poor warnings risk losing the public's trust.

AI companies are pitching chatbots as a fix. Rob Dale, who runs emergency management in Ingham County, Michigan, has used AI to draft practice alerts but says he would still want control over sending one out. Sarah Miller, a professor at Pierce College in Florida, said official adoption is low but unofficial use is real, with officials relying on personal or free accounts.

Not everyone is convinced. Ginny Katz of the consultancy HazAdapt said AI-generated content does not feel authentic, and skeptical residents may spurn agencies that use it. Experts also warn that faster, more frequent alerts could worsen warning fatigue, and that the software is only as capable as the person using it.