Autonomous vehicles, defined as
Currently, fully autonomous taxis are being operated in some major cities in China including Beijing, Shanghai, Guangzhou, Shenzhen, and Wuhan, by tech players including
But this raises a vital question: What party is responsible when an autonomous vehicle (AV) is involved in an incident? Is it the driver, the vehicle manufacturers, or the software developer? How is liability determined?
This has been an
The shift from human-controlled to AI-driven vehicles fundamentally alters the nature of risk and liability. Traditional accident analysis focuses on human error. However, AV incidents often involve complex interactions between software, sensors, mapping data, and the surrounding environment.
Determining fault in AV incidents requires extensive data reporting from components such as data sensor logs, software code, and a vehicle's operating parameters. This rich data will help answer fundamental questions such as: Was the driver or algorithm of each vehicle in control leading up to the impact? Did the driver ignore prompts to take control, or were they texting, playing games, or asleep? If they were prompted, did the system provide sufficient warning to react?
Likewise, conventional underwriting models rooted in driver demographics and historical accident data are no longer adequate. Instead, new models must consider decreased frequency but increased severity of AV-involved accidents and the potential for
Enter data-driven insurance
Drivers and insurers may be reluctant to accept the interpretations of vehicle manufacturers. Plaintiff's lawyers may have great difficulty interpreting the rich streams of visual, lidar, and sensor data captured by the AV leading up to an impact.
In this way, all parties can have a high degree of confidence that relevant information is stored in a "lock-box" that can be readily interpreted by insurers and, if required, the courts in case of a dispute.
In the U.S., recent NHTSA investigations into advanced driver-assistance systems (ADAS) have provided examples of real-world challenges that will accelerate as automated vehicles become more common on the roads. These include incidents involving collisions with emergency vehicles and other scenarios that raised concerns about how AV systems performed in specific scenarios. More transparent reporting on the capabilities and limitations of ADAS technologies would help regulatory bodies develop safety minimums and provide beneficial underwriting parameters for insurers.
AI-powered tools based on the same digital platform enable insurers to develop many dynamic products, such as usage-based insurance (UBI), parametric insurance, and embedded insurance.
Effective data sharing between insurers, automakers, technology providers, and regulators is crucial. These collaborations will enable more accurate risk assessments, faster claims processing, and the development of robust industry-wide best practices.
Conclusion
By leveraging the advancing analytics, fostering cross-industry partnerships, and pioneering efforts to shape regulatory frameworks, insurers can capably navigate the challenges and unlock the immense underwriting opportunity of autonomous vehicles.
Insurers must prioritize investment in data analytics, collaborate with technology providers, and actively engage policymakers to establish clear and consistent standards. By doing so, the industry can build a foundation of trust, safety, and efficiency to propel the autonomous mobility revolution.
