This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Waymo autonomous vehicles in San Francisco's South of Market neighborhood began repeatedly honking at each other while navigating a parking lot used for ride pauses. The behavior, which occurred throughout the night and peaked around 4:00 a.m., caused significant sleep disruption for nearby residents. Waymo acknowledged the issue, attributing it to a glitch in their algorithms, and stated they were working on a fix.
Waymo self-driving cars in San Francisco's South of Market neighborhood reportedly began honking at each other late at night, disturbing residents' sleep. The autonomous vehicles, reportedly using a parking lot for ride pauses, triggered honking due to a purported glitch in their algorithms. Despite residents' complaints, the issue allegedly persisted for weeks until Waymo acknowledged the problem and began working on a fix.
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 Waymo vehicles exhibited a lack of robustness in their navigation and interaction algorithms, leading to inappropriate and repetitive honking behavior in parking lots.
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 disturbance was caused by the autonomous decision-making of deployed Waymo vehicles reacting to each other, which was an unintended outcome of their navigation algorithms.
Added categories peak at 2: Minor (Psychological Harm, Inferred)
An unintentional multi-agent algorithmic glitch caused autonomous Waymo vehicles to repeatedly honk at each other in a San Francisco parking lot, leading to sleep disruption for local residents. The incident represents a minor localized safety and noise issue with negligible implications for national security.
The incident involves multiple autonomous Waymo vehicles interacting with each other in a shared parking lot, where the actions of one vehicle (backing up) triggered automatic honking reactions from the others, creating a collective feedback loop.
Risk Level 2: High Risk. The Waymo Driver is an AI system deployed in critical transport infrastructure, which has significant implications for public safety and public interest.
The primary function of the Waymo Driver system is to operate self-driving vehicles autonomously on public roads and navigate parking lots.
The reports provide clear, consistent, and detailed accounts of the incident from multiple residents and include Waymo's official response acknowledging the issue. The cause (algorithmic behavior during parking lot navigation) and the effects (sleep deprivation for neighbors) are explicitly stated without conflicting details.
None
Technology
Automatic Honking Trigger
Vehicles automatically honk when other vehicles get close while backing up.
Flawed Proximity Logic
Algorithms treat other backing-up Waymos as hazards requiring a horn warning.
Lack of Depot Quiet Mode
Software lacks a quiet mode for vehicles navigating residential parking lots.
Data Inputs
Proximity Sensor Inputs
Sensors detect nearby moving vehicles, triggering default safety honk rules.
Lack of Environmental Context
System lacks data inputs indicating the vehicle is in a private parking lot.
Human Factors
Lack of On-Site Operators
No human drivers present in the vehicles to stop the honking manually.
No Local Contact Person
Neighbors have no direct human contact on-site to address immediate issues.
Process and Methods
Inadequate Depot Testing
Testing processes failed to identify multi-car parking lot interaction issues.
Slow Feedback Loop
Corporate response channels were slow to address neighbor complaints initially.
Management
Inadequate Risk Assessment
Management failed to assess community noise impacts of vehicle depot operations.
Slow Initial Response
Management did not address neighbor complaints until media inquiry.
Regulatory Environment
Lack of AV Noise Regulations
No specific local regulations governing autonomous vehicle noise in depots.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
