Who's at Fault When a Driverless Car Crashes?

Autonomous vehicles are no longer a future concept. They're sharing the road today, and when one is involved in a collision, the fault question gets complicated fast. Was it the driver, the automaker, the software, or a mix of all three? The answer increasingly depends on data most people never see: sensor logs, event data recorders, and a careful reconstruction of what actually happened in the seconds before impact.

Courts are still working out how traditional liability rules apply to a vehicle that can, in part or in full, drive itself. Historically, fault in a crash came down to driver negligence. That framework still applies to most cases today, but claims involving automated systems increasingly raise product liability theories as well, including design defect, manufacturing defect, and failure to warn. Under these theories, a manufacturer or software developer can be held responsible if a flaw in the system, rather than driver error, caused or contributed to the crash. As more vehicles ship with advanced driver-assist and automated features, plaintiffs' attorneys, defense counsel, and insurers are all adjusting how they build and evaluate these claims.

Regulators are paying closer attention too. Since 2021, NHTSA has required manufacturers and operators of automated driving systems and Level 2 driver-assist vehicles to report qualifying crashes under its Standing General Order. The order was updated again in 2025 to streamline reporting and reduce duplication between companies, but the underlying purpose hasn't changed: giving investigators a real-time picture of how these systems perform in actual crashes rather than in controlled testing. That data is increasingly relevant to individual claims as well, since it can establish how a given system has performed across a broader set of incidents.

How Fault Typically Breaks Down

• Level 0-2 driver assistance systems. Features like lane-keeping and adaptive cruise control still require the driver to actively supervise the vehicle. Fault typically stays with the driver unless a component defect is identified.

  
  

• Level 3 conditional automation. The system can handle driving tasks in certain conditions but must hand control back to the driver when prompted. Liability often hinges on whether the handoff warning was adequate and whether the driver responded to it.

  
  

• Level 4-5 high and full automation. The vehicle is designed to operate with little or no human intervention. When something goes wrong, liability shifts toward the manufacturer or software developer, especially when no licensed driver is actively monitoring the vehicle.

  
  

• Third-party contributors. Roadway design, signage, maintenance providers, and mapping data can all factor into a crash independent of how the vehicle's automation performed.

Where Forensic Evidence Comes In

Determining fault in an autonomous vehicle case draws on several sources of physical and digital evidence, including:

• Event data recorder (EDR) output covering speed, braking, and steering input in the moments before impact

• Tire, skidmark, and steering analysis to confirm vehicle behavior at impact

• Biomechanical and medical analysis to correlate injuries with crash dynamics

• Roadway and highway design review, since infrastructure can contribute to a crash independent of the vehicle's automation

• Software and system log review to establish what the vehicle detected, predicted, and decided in real time

Each of these data points needs to be interpreted by someone who understands both the engineering and the legal standard of proof. Data alone doesn't win or lose a case. How it's extracted, preserved, and explained to a judge or jury does.

How We Support Your Case

We provide automotive accident reconstruction and forensic engineering services to insurance adjusters, defense attorneys, and personal injury attorneys nationwide. Our network of engineers and associates handles failure analysis, data extraction, and expert witness testimony, from initial claim review through deposition and trial.

If you're evaluating a claim involving an autonomous or driver-assist vehicle, an early technical review can clarify what the evidence actually supports before positions harden on either side.

Ready to discuss a case? Reach out to our team for a consultation.

---

Disclaimer 

The content provided on this website is for informational and educational purposes only and is not intended as legal, professional, or expert advice. You should consult with a qualified professional for advice tailored to your specific situation or needs. Do not take or refrain from taking action based on the information contained on this site without seeking appropriate professional consultation. The information on this website may not reflect the most up-to-date industry developments or standards. We disclaim all liability for any actions taken or not taken based on the content of this site to the fullest extent permitted by law.

HOME