The safety and laws around self-driving cars

How safe are they?

Updated August 9, 2023

Reviewed By Ross Martin

Table of contents:

The promise of self-driving cars
Levels of automation
Federal and state regulations
Challenges and concerns
Potential solutions
Conclusion

Self-driving cars, once a far-fetched concept reserved for science fiction, are now rapidly becoming a reality. The Insurance Institute for Highway Safety predicts 4.5 million self-driving vehicles on U.S. roads by 2030^[1]. The promise of autonomous vehicles lies in their potential to revolutionize transportation, reducing accidents, alleviating traffic congestion and offering enhanced mobility to people with disabilities or limited driving abilities. The National Highway Traffic Safety Administration shows in a study of the U.S. annual vehicular fatality rate that 94% of crashes are due to human error^[2].

But how safe are self-driving cars? This article explores the key arguments, challenges and potential solutions that shape the discourse on the safety of autonomous vehicles.

The promise of self-driving cars

Proponents of autonomous driving highlight the numerous potential benefits of these driving systems. The most significant advantage is the potential to significantly reduce traffic accidents and fatalities. Human error is the leading cause of most accidents, and self-driving cars, with their advanced sensors and artificial intelligence, have the potential to mitigate these errors, thereby making roads safer for all users. A KMPG report predicts that by 2050, the adoption of self-driving vehicle technology could reduce accidents by almost 90%^[3].

Furthermore, autonomous cars could potentially enhance road efficiency by optimizing traffic flow, reducing congestion and minimizing fuel consumption.

Levels of automation

There is a range of automation technology available in vehicles today. The functionality ranges from vehicle safety measures like emergency braking and blind spot detection to driver assistance and autopilot features.

Automated driving systems are divided into levels:

Level 1:

Consistent driver assistance. This is similar to the technology above but for longer periods of time as opposed to momentary nudges. Examples include adaptive cruise control that can change speed or lane keeping assistance.
Level 2:

Additional assistance. With these technologies, you are still in control of the vehicle, but the automation is capable of both breaking and steering autonomously.
Level 3:

Conditional automation. The above technologies all require the driver to do the majority of the driving, but this level is able to handle all aspects of driving when engaged.
Level 4:

High automation. With this level of driver assistance system the system itself is fully responsible for driving tasks and people in the car are mostly overseeing.
Level 5:

Full automation. At this level, the vehicle is fully a driverless car and occupants are passengers who do not need to be engaged.

Federal and state regulations

At present, not all states allow all levels of automated vehicles on public roads. Federal regulation for self-driving technology is still in the works.

Twenty-nine states — Alabama, Arkansas, California, Colorado, Connecticut, Florida, Georgia, Illinois, Indiana, Kentucky, Louisiana, Maine, Michigan, Mississippi, Nebraska, New York, Nevada, North Carolina, North Dakota, Oregon, Pennsylvania, South Carolina, Tennessee, Texas, Utah, Virginia, Vermont, Washington and Wisconsin —and Washington D.C. have enacted legislation related to autonomous vehicles^[4].

Not all of these states are the same. Only seven states (Florida, Georgia, Nevada, North Carolina, North Dakota, Utah and West Virginia) do not require a driver behind the steering wheel for all Level 4 and Level 5 self-driving cars.

Challenges and concerns

Despite the promises, self-driving cars are not without their detractors, which have fueled the safety debate. Here are some of the most pressing issues:

Technical limitations: Self-driving cars rely on complex algorithms and sensors to navigate, and these systems are not perfect. Adverse weather conditions, unclear lane markings and unexpected obstacles can still pose challenges for autonomous vehicles.
Cybersecurity: As self-driving cars become more interconnected, they may become vulnerable to hacking or malicious attacks. Breaches in cybersecurity could have catastrophic consequences, endangering lives and undermining public trust in the technology.
Ethical decisions: Self-driving cars may face situations where they need to make split-second decisions that involve weighing potential harm to passengers, pedestrians or other road users. The debate around the ethical implications of such decisions remains unresolved.
Transition phase: During the transition from human-driven to fully autonomous vehicles, a mixed environment on the freeways may introduce new risks. The unpredictability of human drivers around self-driving cars could lead to accidents.
Legal and regulatory framework: The lack of consistent and comprehensive regulations governing self-driving cars has slowed their widespread deployment. The absence of clear guidelines and safety standards raises questions about accountability in the event of accidents or system failures.

Potential solutions

Addressing the safety concerns around self-driving cars requires a multi-faceted approach involving various stakeholders:

Rigorous testing and validation: Car manufacturers and technology companies must conduct extensive real-world testing under diverse conditions to identify and rectify potential safety issues. Collaboration with researchers, regulators and third-party organizations can enhance the credibility of the testing process.
Improved sensor technology: Advancements in sensor technology are crucial to enhancing the accuracy and reliability of self-driving cars. Continued research and development will be necessary to address challenges like inclement weather and difficult road conditions.
Cybersecurity measures: Robust cybersecurity protocols and encryption techniques must be integrated into the design of self-driving cars to safeguard against potential cyber threats.
Ethical frameworks: The development of universally accepted ethical frameworks can guide the decision-making processes of self-driving cars in critical situations. Public input and consensus will be essential in shaping these frameworks.
Public awareness and education: Educating the public about the technology and its limitations can help manage expectations and foster acceptance. Raising awareness of the potential safety benefits and the ongoing efforts to address concerns can build trust in self-driving cars.

How will self-driving cars impact car insurance?

Self-driving cars bring up a number of interesting issues related to car insurance. Who is liable in the event of a crash? How will it impact your driving record? Read our predictions on self-driving cars and car insurance.

Conclusion

The safety debate surrounding self-driving cars reflects the transformative potential of this technology and the complex challenges it presents.

While advocates argue that autonomous vehicles could dramatically reduce accidents and reshape the future of transportation, skeptics raise valid concerns about technical limitations, cybersecurity risks and ethical dilemmas.

Striking the right balance between innovation and safety will require collaboration between technology developers, regulators, researchers and the public. As self-driving cars continue to evolve, finding solutions to these safety challenges will be critical in determining the success of this revolutionary mode of transportation.

Sources