This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A federal judge upheld a $243 million jury verdict holding Tesla partially liable for a fatal 2019 crash in Florida where its Autopilot system failed to brake or warn the driver before striking pedestrians.
On April 25, 2019, George McGee's Tesla Model S, with Autopilot reportedly engaged, drove through a Key Largo intersection and struck a parked Chevrolet Tahoe, which then hit two people standing beside it: Naibel Benavides Leon, who was killed, and Dillon Angulo, who was seriously injured. McGee admitted distraction. In 2025, a federal jury found Tesla 33% responsible on defective-design and failure-to-warn claims. Tesla disputed liability and said it would appeal.
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
Tesla's marketing and system design fostered driver complacency and overreliance, leading the driver to trust the system to avoid crashes while he looked away.
The primary failure lies in the Autopilot system's lack of capability and robustness to handle a T-intersection, detect stationary profiles, and execute emergency braking or collision warnings despite registering the obstacles.
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 incident was caused by the failure of Tesla's Autopilot system to brake or warn the driver after detecting obstacles, which was an unexpected and unintentional outcome of its operation post-deployment.
Added categories peak at 2: Minor (Psychological Harm, Direct)
A fatal 2019 vehicle crash involving Tesla Autopilot led to a $243 million civil judgment in 2026, highlighting persistent safety and liability challenges in consumer driver-assist technologies. While the incident carries significant implications for corporate liability and transportation safety regulations, its direct national security impact remains minor.
The Autopilot system operates independently to control lateral and longitudinal movement of the vehicle in real-time based on environmental inputs, without explicit human planning for each maneuver.
High Risk: The system is an AI application used in critical safety infrastructure and transport (autonomous driving/assisted driving systems), which has significant implications for physical safety and human life.
The primary function of Tesla's Autopilot suite is to assist with or execute autonomous driving tasks, including lane centering, speed adjustment, and emergency braking.
The reports provide highly detailed, consistent, and legally verified accounts of the crash, the technical failures of the Autopilot system, and the subsequent court rulings and forensic data extraction.
Tesla argued that the sole cause of the crash was driver recklessness, but the jury and judge rejected this, finding joint liability.
Technology
Defective Camera-Only Perception
Tesla relied solely on cameras, which struggled with stationary objects.
Failure of AEB and FCW Systems
The emergency braking and collision warnings failed to trigger for the hazard.
Absence of Geofencing Safeguards
The system lacked technical limits to prevent activation on unsafe local roads.
Data Inputs
Inaccessible Crash Snapshot Data
Crucial telemetry and video logs were withheld or claimed corrupted by Tesla.
Unreliable Sensor Redundancy
Lack of LiDAR or radar inputs left cameras vulnerable to poor road geometries.
Human Factors
Driver Complacency and Distraction
The driver was looking down for a dropped phone, expecting system intervention.
Inadequate Driver Monitoring
The system failed to ensure the driver remained actively engaged and attentive.
Overreliance on Autopilot
The driver trusted the technology to automatically avoid obstacles.
Process and Methods
Deceptive Marketing Terminology
Terms like 'Autopilot' and 'Full Self-Driving' fostered a false sense of safety.
Inadequate Display Warnings
On-screen alerts acted as instructions rather than warning of fatal risks.
Unsafe Data Extraction Procedures
Non-forensic handling of the computer threatened to overwrite critical logs.
Management
Prioritizing Sales Over Safety
Tesla chose not to geofence Autopilot to make vehicles more appealing to buyers.
Failure to Act on Safety Advice
Tesla ignored NTSB recommendations to restrict Autopilot to highways.
Lack of Data Transparency
Tesla hid or deleted critical crash data to limit corporate liability.
Regulatory Environment
Lack of Strict ADAS Standards
Regulators did not mandate geofencing or robust driver monitoring in 2019.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
