This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Two self-driving food delivery robots operated by different companies crashed into glass CTA bus shelters in Chicago within a week, causing property damage and fueling public safety concerns.
A Coco delivery robot reportedly collided with a glass bus shelter in Chicago's Old Town neighborhood, shattering panels and scattering glass across the sidewalk. Coco called the crash a rare incident, said no one was injured, and accepted responsibility for cleanup and repair costs while investigating what caused the robot to strike the structure.
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.
The robots failed to perform reliably under real-world conditions, specifically failing to detect and avoid glass bus shelters, leading to collisions.
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 property damage was caused by the autonomous navigation failures of deployed delivery robots operating in public spaces.
Two autonomous delivery robots experienced navigation failures and crashed into public transit shelters in Chicago, causing minor localized property damage. The incident represents a minor national security concern, highlighting localized safety and regulatory challenges of integrating autonomous systems into public spaces rather than a systemic threat.
The self-driving robots operate independently on public sidewalks, making real-time navigation decisions based on environmental inputs without active human planning or direct control.
High Risk: The AI systems operate in public critical infrastructure (sidewalks and transit areas) and pose physical safety risks to the public.
The AI systems are self-driving delivery robots designed to autonomously navigate city sidewalks to deliver food.
The report is from a local news outlet detailing two specific, verified crashes with statements from both robot companies and city officials. The facts of the collisions and the lack of injuries are clear.
None.
Technology
Glass Detection Failure
Robots from both companies failed to detect transparent glass structures.
Navigation System Error
Autonomous systems directed the robots directly into stationary obstacles.
Data Inputs
Sensor Input Limitations
Optical or lidar sensors failed to register transparent glass barriers.
Human Factors
Lack of Operator Intervention
Remote monitors did not override the autopilot to prevent the collisions.
Process and Methods
Insufficient Sidewalk Testing
Testing processes did not account for common transparent sidewalk hazards.
Management
Premature Public Deployment
Management deployed robots in public areas before resolving collision risks.
Lack of Public Transparency
Companies failed to provide immediate explanations for the safety failures.
Regulatory Environment
Inadequate Pilot Oversight
The city pilot allowed robot deployment without strict safety validation.
