This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
On June 3, 2022, a Cruise autonomous vehicle operating in driverless mode was involved in a collision at an intersection in San Francisco. The vehicle attempted an unprotected left turn, stopped in the intersection due to an oncoming speeding Toyota Prius, and was subsequently struck by the Prius. Occupants in both vehicles sustained minor injuries, and the incident led to a voluntary software recall by Cruise to improve predictive performance in similar scenarios.
A Cruise autonomous vehicle was involved in a crash at an intersection in San Francisco when making a left turn in front of a Toyota Prius traveling in an opposite direction, which caused occupants in both cars to sustain injuries.
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 Cruise AV failed to perform robustly under complex real-world conditions, specifically failing to correctly predict and react to the sudden path change of a speeding vehicle violating lane usage.
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 collision was caused by the Cruise AV's decision to hard brake and stop in the intersection, which was an unexpected outcome of its reflexive planner software attempting to avoid a collision.
A Cruise autonomous vehicle operating in driverless mode was involved in a collision in San Francisco, resulting in two minor injuries and a subsequent voluntary software recall. This represents a minor physical safety failure in public transportation infrastructure with negligible broader national security implications.
The Cruise AV was operating independently in 'driverless autonomous mode' on public roads, making real-time driving decisions and reacting to environmental inputs (the oncoming Prius) without human intervention.
High Risk: The system is an autonomous vehicle operating in public traffic infrastructure, which acts as a safety component of a transport system and poses significant risks to physical safety.
The primary function of the Cruise ADS is to operate a vehicle autonomously on public roads, navigating traffic and making real-time driving decisions.
The reports provide highly detailed and consistent accounts of the crash, including official regulatory filings from Cruise to the NHTSA and DMV that explain the exact software mechanics (reflexive planner) and the sequence of events.
None, the technical cause and physical outcomes of the crash are well-documented and agreed upon by the reporting sources.
Technology
Reflexive Planner Hard Braking
The reflexive planner caused the AV to hard brake and stop mid-turn.
Inadequate Prediction Software
The ADS failed to correctly predict the Prius's sudden straight trajectory.
Inability to Negotiate Complex UPLs
Unprotected left turns are highly complex and prone to planning errors.
Data Inputs
Incorrect Lane Usage Assumption
The AV assumed the Prius would turn right because it was in a turn lane.
Speeding Vehicle Detection Lag
The AV struggled to react safely to a vehicle traveling 40 mph in a 25 zone.
Human Factors
Other Driver Speeding
The Prius driver was traveling at 40 mph in a 25 mph speed zone.
Other Driver Lane Violation
The Prius driver went straight from a designated right-turn/bus lane.
Lack of Human-AV Communication
No effective way for the AV and human driver to communicate intentions.
Process and Methods
Inadequate Edge Case Testing
The system had not fully resolved the rare reflexive planner UPL bug.
Management
Prioritizing Commercial Scale
Management pushed for commercial expansion despite known edge-case risks.
Internal Safety Concerns Unresolved
An anonymous employee alleged unsafe practices and pressure to scale up.
Regulatory Environment
No Federal AV Performance Standards
NHTSA had no specific performance standards for fully autonomous vehicles.
Premature Commercial Permits
Regulators granted a commercial fare permit just one day prior to the crash.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
