Cruise autonomous vehicles experienced multiple system outages in San Francisco where nearly 60 vehicles lost connection to company servers and stopped functioning, blocking traffic and trapping other drivers for extended periods.
Between May and June 2022, Cruise autonomous vehicles operated by General Motors experienced several system outages in San Francisco that caused significant traffic disruptions. On May 18, the company lost contact with its entire fleet for 20 minutes, leaving cars stopped in streets with no ability to communicate with passengers or remotely steer vehicles to safety. On June 28, nearly 60 vehicles were disabled across the city over a 90-minute period after losing connection to Cruise servers, with as many as 20 cars creating traffic jams downtown and some stopping in crosswalks. During one incident near Golden Gate Park, driver Calvin Hu and other motorists including a paratransit bus were trapped in a 'robotaxi sandwich' for at least 15 minutes, forcing Hu to drive over curbs to escape. An anonymous Cruise employee reported to the California Public Utilities Commission that the company loses contact with vehicles 'with regularity,' sometimes requiring tow trucks for recovery. The incidents occurred during a critical expansion period for Cruise, which had recently received permits to charge for rides and was competing with other autonomous vehicle companies. Company spokesperson Tiffany Testo stated that vehicles are programmed to pull over and activate hazard lights during technical problems, calling such incidents 'one aspect of our overall safety operations.'
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.
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