The rise of autonomous and semi-autonomous vehicles has reshaped conversations about road safety, technology, and accident liability. Once confined to experimental labs and closed tracks, self-driving cars are now operating in cities across the United States and globally. While these vehicles promise fewer human-error accidents, real-world crash data is essential for understanding actual performance and safety risks.

Drawing on credible sources like the National Highway Traffic Safety Administration (NHTSA), we’re here to break down self-driving car accident statistics from 2019 to 2024. In this guide, you’ll read about trends, company-specific data, common causes, and key safety insights — all underscoring the importance of legal guidance when navigating the aftermath of self-driving car accidents.

Table of Contents

Key Takeaways — Self-Driving Car Crash Trends (2019–2024)

More than 800 reported crashes involving self-driving or advanced driver-assistance systems (ADAS) since NHTSA began mandatory reporting in 2021.
Tesla vehicles account for the majority of Level 2 ADAS crash reports, followed by Honda and Subaru.
California, Texas, and Arizona lead in reported crashes involving vehicles equipped with automated driving systems (ADS). This statistic reflects each state’s high testing and deployment rates.
Fatalities remain rare compared to total incidents, but injury crashes are increasing as adoption grows.
Common crash types include rear-end collisions, lane-change accidents, and interactions with pedestrians or cyclists.

Source: NHTSA, Standing General Order on Crash Reporting

What Counts as a Self-Driving Car Accident?

Understanding the terminology is crucial before diving into the numbers. The media often uses “self-driving” loosely, but in safety reporting, the definitions are precise:

Levels of Vehicle Automation

Level 0 – Some automation; human driver is fully engaged and vehicle assists with features such as lane departure warning and forward collision warning.
Level 1 – Driver assistance for a single task (e.g., adaptive cruise control).
Level 2 – Partial automation; vehicle can control steering and speed simultaneously, but driver must remain engaged.
Level 3 – Conditional automation; vehicle handles most tasks but requires driver intervention when requested.
Level 4 – High automation; vehicle operates without human input in specific conditions or geofenced areas.
Level 5 – Full automation; no human driver needed in any conditions, and the vehicle is fully responsible for driving tasks.

Source: NHTSA, Automated Vehicles for Safety

ADS vs. ADAS Systems

Automated Driving Systems (ADS) typically refer to Level 3–5 technology, capable of fully controlling the vehicle in certain scenarios.
Advanced Driver-Assistance Systems (ADAS) are Level 1–2 features like lane-keeping assist or adaptive cruise control, which support — but don’t replace — the driver.

Reporting Requirements & Data Limitations

NHTSA’s Standing General Order Crash Reporting mandates that manufacturers and operators report certain crashes involving ADS and Level 2 ADAS within 24 hours. However, the data has limitations:

Only incidents meeting specific injury or vehicle damage thresholds are reported.
Some manufacturers provide more complete data than others.
Early adoption areas may appear riskier due to more extensive testing and reporting.

Source: NHTSA, Standing General Order on Crash Reporting

How Many Self-Driving Car Accidents Happen Each Year?

NHTSA’s reporting system began compiling national ADAS and self-driving car accident data in June 2021, though some state-level and company-reported data extend back to 2019. Trends show steady increases, driven by greater deployment, expanded testing zones, and improved reporting compliance.

Year-by-Year Crash Estimates

2019 – Fewer than 50 publicly known incidents, mostly from pilot programs in California and Arizona.
2020 – Activity slowed due to pandemic restrictions; around 60 reported incidents.
2021 – Over 200 incidents after NHTSA’s reporting mandate began mid-year.
2022 – Approximately 400 incidents, with more manufacturers submitting complete data sets.
2023 – Nearly 500 incidents were reported, including several high-profile collisions in San Francisco and Phoenix.
2024 – Preliminary data suggests another rise, partly due to expanded autonomous taxi fleets and more public road testing.

Source: NHTSA, Standing General Order on Crash Reporting

Why Certain Years Spike

Fleet growth: More automated vehicles (AVs) on the road increase exposure to potential incidents.
Improved reporting: Early years suffered from underreporting, making later figures appear larger.
Urban deployment: Concentrated operations in dense cities lead to more collision opportunities.

Self-Driving Car Accidents by Company

Self-driving car accident counts often reflect both the size of a company’s fleet and how aggressively it deploys automation. Let’s break down self-driving car crash data by manufacturer:

Company	System Type	Reported Crashes	Notes
Tesla	Level 2 ADAS	2,000+	Large fleet, nationwide; many driver-assist crashes involve inattentive drivers.
Honda	Level 2 ADAS	100+	Primarily lane-keeping and adaptive cruise systems.
Subaru	Level 2 ADAS	50+	EyeSight driver-assistance system widely deployed.
Waymo	Level 4 ADS	1,000+	Operating fully driverless rides in Phoenix, and limited San Francisco zones.
Cruise LLC	Level 4 ADS	150+	Notable incidents in dense urban environments like SF.
Other OEMs	Mixed	50+ combined	Includes Ford, Toyota, and Mercedes-Benz test programs.

Key Insight:
A high number of crashes for companies like Tesla doesn’t necessarily mean the technology is riskier — larger fleets and nationwide coverage naturally yield more incidents. Additionally, counts for ADS companies like Waymo reflect narrower operational zones.

Source: NHTSA, Standing General Order on Crash Reporting

Geographic Breakdown — Where Do Self-Driving Car Crashes Happen Most?

Self-driving car accidents cluster in states with permissive testing laws, dense urban environments, and favorable weather conditions for AV sensors.

Top States for AV Crashes

California – San Francisco, Los Angeles, and San Jose see heavy AV testing and rideshare operations.
Texas – Austin and Houston have growing autonomous delivery and taxi programs.
Arizona – Phoenix metro is a long-standing testing hub for Waymo and other AV companies.

City-Level Hotspots

San Francisco, CA – Complex road layouts and dense traffic produce frequent AV event reports.
Phoenix, AZ – Long-time AV proving ground with wide roads and predictable weather.
Austin, TX – Emerging hub with mixed traffic types and active AV pilot programs.

Source: NHTSA, Standing General Order on Crash Reporting

Injuries and Fatalities in Autonomous Vehicle Crashes

While self-driving car accidents remain a small fraction of total U.S. crashes, the severity of incidents is a key safety measure. NHTSA reports indicate:

Injuries – Since 2021, more than 2,000 injuries have been linked to self-driving or ADAS-involved crashes. Most were minor to moderate, but dozens required hospitalization.
Fatalities – Just over 50 confirmed fatalities nationwide in the same period.
Severity distribution – Rear-end and low-speed incidents often result in minor injuries, while high-speed or pedestrian-related crashes are more likely to cause serious harm.

Why severity matters: Even if total incidents are relatively low, AV systems must be judged on how well they prevent the most dangerous collisions — not just on the total crash count.

Source: NHTSA, Standing General Order on Crash Reporting

Common Causes of Self-Driving Car Accidents

Legal analyses show recurring crash scenarios:

Proceeding straight – AVs may misjudge speed or distance when approaching intersections or cross traffic.
Stopped or stationary object collisions – Sensor misinterpretation of parked vehicles or road debris.
Turning accidents – Challenges with yielding, gap judgment, or detecting oncoming vehicles.
Lane-change incidents – Blind-spot detection errors or insufficient merge clearance.
Pedestrian and cyclist interactions – Occasional misclassification or delayed recognition in crosswalks.

Are Self-Driving Cars Safer Than Human Drivers?

The core promise of AV technology is reducing human error, which causes over 90% of traditional crashes. However, current comparisons between AV and human driver safety are inconclusive:

Pros

AVs don’t drive impaired, distracted, or fatigued.
Consistent adherence to traffic laws and speed limits.
Potential to reduce high-severity crash types.

Cons

Limited experience with unpredictable human driver behavior.
Technology gaps in adverse weather and complex traffic patterns.
The sample size is still too small for definitive safety comparisons.

Key Insight:

These comparisons suggest that AVs could prevent many crashes caused by perception and attention errors — but only if designed to prioritize safety over trip speed or efficiency.

Public Perception of Autonomous Vehicle Safety

Despite technological advancements, public trust in AV safety remains mixed. Surveys from industry and government sources show:

Roughly half of U.S. adults say they would be uncomfortable riding in a fully self-driving vehicle.
Top concerns: System failure, unclear liability in crashes, and lack of human control.
Younger drivers tend to express greater acceptance, while older demographics remain more skeptical.

Perception often shifts after high-profile incidents, suggesting that transparency and consistent safety performance will be critical to building trust.

Source: PolicyAdvice, 25 Astonishing Self-Driving Car Statistics

Legal Responsibility in Self-Driving Car Accidents

Determining liability in a self-driving car crash can be more complex than in a human-driven accident. Factors include:

Automation level – Higher autonomy often shifts more responsibility to the manufacturer or operator.
Driver monitoring requirements – If the system required a driver’s attention and they failed to intervene, liability may fall partly on them.
Software or hardware defects – May point to product liability claims.
Fleet operation policies – Commercial AV services could be held liable if negligent maintenance or training contributed to the crash.

Potentially Liable Parties

Driver or operator
Manufacturer (vehicle or components)
Software developer
Fleet owner/operator

If you’ve been involved in a self-driving vehicle crash, speak with an experienced car accident attorney for help clarifying your legal options.

Frequently Asked Questions

What’s the difference between ADS and ADAS?

ADS refers to high-level automation (Levels 3–5) capable of full driving tasks under certain conditions. ADAS (Levels 1–2) offers driver-assist features like lane-keeping or adaptive cruise control, but still requires human oversight.

How do authorities investigate AV crashes?

NHTSA collects crash reports from manufacturers, often followed by independent state or federal investigations. Data from vehicle sensors, cameras, and system logs is key evidence.

Can I sue after being hit by a self-driving car?

Yes — depending on fault, you may have claims against the driver, manufacturer, or operator. Legal advice is essential for navigating liability in AV cases.

Are there fully autonomous vehicles on public roads today?

Yes, but only in limited geofenced areas. Companies like Waymo and Cruise operate Level 4 vehicles in select U.S. cities.

The Risks and Realities of Self-Driving Car Accidents

The growth of AV technology brings both promise and challenges. Data shows that while self-driving cars can reduce some risks associated with human drivers, they are not accident-proof. Early crash statistics should be interpreted with caution — fleet sizes, reporting standards, and geographic deployment vary widely. As testing and deployment expand, continued monitoring, transparency, and informed debate will be critical to understanding and improving AV safety.

If you or a loved one has been injured in a self-driving car accident, contact Naqvi Injury Law to schedule a free consultation, learn your rights, and explore potential claims. As an experienced car accident lawyer in Las Vegas, we leverage years of legal advocacy to ensure you receive the maximum compensation possible for your injuries.

Related Content: