Cruise's Self-Driving Cars Allegedly Lost Connection to Their Server, Causing Traffic Blockages in San Francisco

Cruise’s autonomous vehicles were shown on video stopping in the middle of the road and causing blockages in San Francisco, as they were disabled allegedly due to lost connection to their company’s server.

Risk classification

• Primary risk domain: 7 AI system safety, failures, & limitations
• Primary risk subdomain: 7.3 Lack of capability or robustness

The Cruise vehicles failed to perform reliably under real-world conditions, repeatedly freezing and blocking traffic due to server connectivity losses and an inability to handle specific road scenarios.

Additional risk subdomains

• 6.5 Governance failure: The reports highlight regulatory challenges, including a state law loophole preventing moving violation citations for driverless cars and the CPUC approving commercial permits despite safety objections from local fire, police, and transit officials.

Causal factors

• Entity: AI
• Intent: Unintentional
• Timing: Post-deployment

The incidents were caused by the Cruise autonomous vehicles experiencing unexpected technical failures and server connectivity losses after being deployed on public roads.

EU AI Act risk tier

• Risk tier: 2 High Risk

High Risk: The AI system is deployed in critical road transport infrastructure. The reports describe vehicles blocking active traffic lanes and obstructing emergency services, which directly impacts public safety.

AI system and alleged parties

• AI system: Cruise autonomous vehicles (General Motors)
• AI purpose: Autonomous Driving; Navigation Assistant
• Behaviour type: Autonomous
• Alleged developer: Cruise
• Alleged deployer: Cruise
• Alleged harmed parties: San Francisco traffic participants, San Francisco public

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Minor. Severity is scored from Negligible to Catastrophic in each harm category, for harm the reports describe as caused directly or indirectly by the AI system.

• Physical: direct Negligible, indirect Negligible
• Infrastructure: direct Minor, indirect Minor
• Property: direct Negligible, indirect Negligible
• Financial: direct Negligible, indirect Negligible
• Environmental: direct Negligible, indirect Negligible
• Malicious content: direct Negligible, indirect Negligible
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Negligible, indirect Negligible
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Negligible, indirect Negligible
• Epistemic: direct Negligible, indirect Negligible
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Infrastructure

Reported: The report explicitly describes temporary infrastructure disruption caused directly by the incidents.

Directly caused: Cruise vehicles blocked multiple active traffic lanes on Gough, Fulton, Fell, and Masonic streets, rendering key public roadways impassable for hours.

Indirectly caused: The immobilized vehicles disrupted public transit routes (trapping a paratransit bus) and prevented regular municipal street sweeping operations.

People affected

• Occurrences reported: 5
• People reportedly harmed: 50
• People reportedly exposed: 200

Potential causes

Management

• Prioritizing Commercialization: Rushing to launch paid services to compete with other autonomous vehicle firms.
• Downplaying Systemic Failures: Management dismissed frequent fleet freezes as standard safety operations.
• Inadequate Risk Assessment: Management failed to address known regular contact loss before scaling fleet.

Technology

• Loss of Server Connection: Vehicles lost contact with central server, triggering a safety freeze.
• Inaccessible Backup System: Remote operators could not access fallback system to steer stopped cars.
• Inability to Pull Over: Vehicles were not technologically capable of pulling over to the curb.

Data Inputs

• Loss of Server Telemetry: Loss of connection cut off the essential data stream from central server.
• Lack of Vehicle Location Data: Staff could not see where vehicles were located during the outage.
• Lack of Passenger Communication: Staff could not communicate with passengers inside vehicles during outages.

Human Factors

• Delayed Manual Recovery: Staff had to manually drive each vehicle to resolve physical blockades.
• Whistleblower Safety Concerns: An employee raised concerns about regular contact loss and safety risks.
• Emergency Responder Frustration: SFFD was highly concerned about response times blocked by stopped vehicles.

Process and Methods

• Fail-Safe Defaults to Stopping: Vehicles are programmed to stop in place when encountering technical issues.
• Ineffective Remote Operations: Fallback processes failed to let remote operators steer stopped vehicles.
• Inadequate Incident Reporting: Cruise did not publicize the frequency of these fleet-wide outages.

Regulatory Environment

• Premature Commercial Approval: CPUC approved driverless operations despite warnings from city officials.
• Moving Violation Loophole: State law prevents police from citing driverless vehicles for moving violations.
• Lack of Safety Enforcement: CPUC allowed operations to continue despite knowing about the blockages.

Information quality

• Classification confidence: High
• Reason for confidence: The analysis is supported by multiple detailed journalistic reports (Wired, SF Examiner, SFist) that cite internal Cruise communications, eyewitness accounts, and official letters from San Francisco city departments.
• Ambiguities identified: The exact technical root cause of the server connectivity loss is not fully detailed in the public reports.
• Alternative interpretations: None, as the failure of the autonomous driving system and its real-world impact are clearly documented.

Multiple Cruise autonomous vehicles experienced server connectivity failures in San Francisco, leading to fleet immobilization that blocked active traffic lanes and obstructed an emergency vehicle. While highlighting safety vulnerabilities in autonomous transit systems, the incident has minor national security implications, limited to localized critical infrastructure disruption.

• Overall national security impact: Minor
• Scope: Single nation
• Primary target: United States
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Long-term. The incidents represent a past event and an ongoing strategic concern regarding the safety and resilience of autonomous vehicle integration into public infrastructure.
• Autonomy: Full autonomy. The vehicles operated in driverless mode, making real-time navigation and driving decisions autonomously. The remote assistance fallback systems failed during the connectivity loss.
• Novelty: Evolved capability. While autonomous vehicle testing is established, the simultaneous immobilization of a fleet due to server connectivity loss represents an evolved capability threat to urban transit.

Impact by dimension

• Physical security: Minor. Multiple Cruise autonomous vehicles lost connectivity, freezing and blocking active traffic lanes in San Francisco. This disrupted local road infrastructure and briefly obstructed an emergency fire truck responding to a call, demonstrating minor localized physical security risks.
• Information security: Negligible. No evidence of information warfare, intelligence compromise, or systematic information manipulation associated with these technical connectivity failures.
• Sovereignty: Negligible. No threat to state authority, constitutional processes, or core government operations. Disruption was limited to local municipal services like street sweeping.
• Economic security: Negligible. While the incidents caused minor local economic delays for commuters and recovery costs for Cruise, they posed no threat to national economic stability or strategic technology assets.
• Societal stability: Negligible. The incident caused localized commuter frustration but did not involve mass surveillance, algorithmic oppression, or threats to civil liberties and societal stability.