This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Navigation apps including Waze and Google Maps have been accused of causing significant public safety issues by rerouting heavy traffic onto narrow residential streets in Los Angeles. This displacement has led to gridlock, blocked emergency vehicle access, and increased accident rates. Furthermore, reports indicate that these apps have directed drivers into active wildfire evacuation zones, prompting city officials to seek legal and regulatory interventions.
For years, Waze has, in an attempt to cut travel times, allegedly caused more traffic and guided drivers to make unsafe and often un-permitted traffic decisions, which was described by a Los Angeles city council member as a threat to public safety.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
AI systems that fail to perform reliably or effectively under varying conditions, exposing them to errors and failures that can have significant consequences, especially in critical applications or areas that require moral reasoning.
Also classified as
Drivers overrelied on the navigation apps during critical wildfire evacuations, leading them into active danger zones.
The massive influx of gridlocked traffic on narrow residential streets caused localized air pollution and exhaust-choked congestion.
The navigation systems failed to perform robustly under varying conditions, such as failing to account for physical street capacity limits and active wildfire evacuation zones.
AI system
Due to a decision or action made by an AI system
Unintentional
Due to an unexpected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The navigation algorithms unintentionally routed heavy traffic onto ill-equipped residential streets and into active wildfire zones post-deployment.
Added categories peak at 2: Minor (Psychological Harm, Indirect)
Commercial navigation apps (Waze and Google Maps) unintentionally disrupted local emergency services and routed drivers into active wildfire zones in Los Angeles due to algorithmic routing failures. While presenting notable local public safety and municipal management challenges, the incident has minor national security implications.
The navigation apps operate independently, responding to real-time environmental inputs (user GPS data, reports) using learned routing patterns to make real-time decisions.
High Risk: The system is used in critical infrastructure (traffic management and routing), and the report describes navigation apps guiding drivers into evacuation areas and causing threats to public safety.
The primary function of Waze and Google Maps is to act as a navigation assistant, routing drivers to their destinations. It also acts to allocate traffic resources across public roads.
The reports provide consistent, detailed accounts from multiple cities, municipal officials, and residents regarding the traffic displacement and safety hazards caused by Waze and Google Maps. The role of the AI routing algorithms is explicitly discussed.
The traffic could be viewed purely as a municipal planning failure rather than an AI failure, but the reports strongly focus on the algorithmic routing as the catalyst.
Technology
Lack of Closed-Loop Prediction
Algorithms do not predict how users will react to routing suggestions.
Flawed Routing Optimization
Algorithm prioritizes individual travel times over street capacity.
No Evacuation Zone Detection
App guided drivers into active wildfire evacuation areas.
Data Inputs
Vulnerability to False Reports
Users could submit fake traffic reports to trick the routing software.
Lack of Street Capacity Data
Algorithm lacks input on physical road capacity and design limitations.
Delayed Real-Time Road Updates
Failure to integrate emergency road closures dynamically.
Human Factors
Driver Over-Reliance on Apps
Drivers follow app instructions down unsafe and unfamiliar streets.
Frustrated Driver Behavior
Increased traffic led to road rage and unsafe driving.
User Manipulation of Software
Residents reported phantom wrecks to deter traffic.
Process and Methods
Inadequate Data Sharing
Poor coordination between cities and app developers.
Lack of Local Traffic Modeling
Routing methods fail to model neighborhood impacts.
Management
Prioritization of User Speed
Management focuses on speed and growth over public safety.
Lack of Collaborative Response
Developers ignored city requests to address hazards.
Regulatory Environment
Absence of App Regulation
No regulatory framework governs routing algorithms.
Jurisdictional Limits
City officials lack jurisdiction to force developer compliance.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
