Determining Legal Liability for Autonomous Vehicles


Major technological advances have occurred over the past decade concerning autonomous vehicles, but there is widespread concern that the rate of development is hampered by uncertainty about the liability of manufacturers for a crash. The patchwork of laws and the resultant uncertainty have prompted calls for the federal safety regulation of autonomous vehicles. The uncertainty is stemmed out of the inevitable result of trying to predict how tort rules governing old technologies will apply to the new and more advanced technology of automated driving.[1] The benefits of autonomous vehicles (AVs) are widely acknowledged, but there are concerns about which party is  to bear responsibility for damages stemming from the use of  AVs which this paper seeks to address.


Modern information technologies and the advent of machines powered by artificial intelligence (AI) has already strongly influenced the world of work in the 21st century.  Artificial intelligence describes the work processes of machines that would require intelligence if performed by humans. The term ‘artificial intelligence’ thus means investigating intelligent problem-solving behaviour and creating intelligent computer systems.[2]

A major project, in which not only the international automotive industry but also the large data processing companies are involved, is autonomous driving.[3] According to the most general definition, an autonomous vehicle (AV) is such a vehicle that can guide itself without human conduction. It is a technical unit which fulfils certain tasks without being dependent on regular human commands.[4] In addition to liberating humans from the task of driving, the technology will cause a migration from private car ownership to commercial car-sharing services, alter the dynamics and underlying infrastructures of urban and suburban living, and most importantly for present purposes substantially reduce the carnage on our roadways.[5] Autonomous vehicles would not eliminate all crashes, but they would significantly enhance motor vehicle safety[6] and will save lives and prevent many more injuries; making a compelling safety case for policies that foster the widespread deployment of this technology. Though the everyday use of autonomous vehicles might seem futuristic, prognoses predict their wide use in the near future.

The world of autonomous and connected vehicles is rapidly evolving. As autonomous technology gradually erodes driver control, the law must be altered in its code and its implementation. It is a significant challenge, but not an insurmountable one and the most important and considerable issue is the liability of autonomous vehicles.


As the automobile market moves towards more automated vehicles, the picture may change in some ways, companies are progressively increasing the extent to which their vehicles are automated. Developers of autonomous technology estimate that autonomous vehicles could reduce traffic fatalities by 90%[7] and it may also have the potential to save energy and reduce air pollution from transportation through efficiency and by supporting vehicle electrification. The shift to AVs should produce a substantial reduction in total personal injury costs associated with auto accidents, including medical expenses, lost income, and of course lost lives. That shift too can be expected to affect the number of insurance payouts.

The National Highway Traffic Safety Administration’s (NHTSA) 2013 Preliminary Statement established five levels of vehicle automation to track the progression of vehicles.[8] Level 0 vehicles have no automation and the driver is in “complete and sole control of the primary vehicle controls (brake, steering, throttle, and motive power) at all times and is solely responsible for monitoring the roadway and for the safe operation of all vehicle controls.”[9] At a slightly more automated level, Level 1 vehicles have function-specific automation such as adaptive cruise control, electronic stability control, or dynamic brake support. In limited situations, these technologies can assume control of the car or assist with control. Level 2 vehicles have multiple automated functions that work together such that the vehicle can assume active control of the vehicle in limited driving situations. Tesla’s ‘autopilot’ feature has been described as an NHTSA Level 2 technology feature. [10] Level 2 vehicles, such as Tesla’s Model S, may use a combination of forward-looking radar, outward-facing cameras, sensors, and GPS to visualize the road. Level 3 vehicles offer limited self-driving automation, enabling the driver to cede full control of all safety-critical functions under certain traffic or environmental conditions and in those conditions to rely heavily on the vehicle to monitor for changes in those conditions requiring transition back to driver control. The driver must be able to monitor the vehicle and be capable of assuming control of the vehicle in situations where the vehicle is no longer able to support automation. In Level 4, the vehicle is intended to perform all safety-critical driving functions and monitor roadway conditions for an entire trip. Level 5 AVs are fully automated AVs that take control of all driver functions, the automated driving system (ADS) on the vehicle can do all the driving in all circumstances. The human occupants are just passengers and need never be involved in driving.[11]


Innovative technologies such as AVs create risks and unintended consequences that may decrease society’s acceptance of them. One of the issues of greatest concern to parties involved in the production and use of AVs is how the risks of automobile accidents will be allocated. Before the use of automated vehicles is adopted widely, this question of liability is a matter that should be on the agenda of the Parliament in each state or country, ideally to develop a uniform international response addressing these issues. Tort law does not necessarily apply to new technologies in an easily predictable or wholly rational manner,[12] any framework for automated vehicle operations should include procedurally swift, financially feasible and fair avenues of appeal and redress for users, passengers, owners, purchasers, operators and repairers.[13]

Presumably, when an accident occurs involving an AV, the same negligence elements apply to the operators/passengers in charge of the AVs.  However, AVs are more complicated, and the rules surrounding them remain uncharted. Plaintiffs often seek causes of action under negligence, breach of contract or warranty, fraud, or misrepresentation theories however, there is a need for a more thorough form of liability. Most commentators agree that as vehicles become increasingly autonomous, the liability associated with vehicular accidents will shift from the driver to those involved in deploying the technology.[14] Manufacturers might also face liability under products liability theories. The types of tort claims that can be brought against automotive defendants (manufacturers, distributors, part suppliers, etc.) generally include negligence claims and product liability claims.


Product liability is the branch of law in which manufacturers, distributors, suppliers and others who make products available to the public are held responsible for the injuries those products cause.[15] This section addresses how tort law (especially products liability law) might be applied to losses caused by AV-related crashes.

 Today, all but a few jurisdictions have some version of products liability law. Although there are some variations across jurisdictions, in general, to establish a products liability claim against an automobile manufacturer or any seller of an automobile, a plaintiff must establish three elements. First, the plaintiff must prove that the automobile, or some aspect of the automobile, was “defective.” A car or some part of the car can be defective concerning how it is manufactured, how it is designed, and/or whether there are adequate warnings or instructions regarding particular risks. On the design defect question, most courts apply a standard that focuses on whether the foreseeable risks of harm posed by the product could have been reduced or avoided by the adoption of a reasonable alternative design.[16] This design-defect standard is similar to the negligence standard of reasonable care as applied by some courts. The second element in a products liability claim against an automobile seller is that the defect in the automobile caused the plaintiff’s losses. This showing of causation includes both a showing that the automotive defect was the “but for” cause of the losses (but for the defect, the loss would not have happened) and a showing that the defect was the proximate cause of the losses, that the harm was a reasonably foreseeable result of the defect. Plaintiffs can attempt to show that the defect caused the accident, crash or that it caused the vehicle to be insufficiently “crashworthy,” resulting in harms that would not otherwise have happened.[17] Finally, plaintiffs must prove the nature and extent of the harms they suffered as a result of the automotive defect.[18] Damages that can be recovered include compensatory damages and, in most jurisdictions, punitive damages.

Even if a plaintiff can establish the preceding elements of a products liability claim against an automotive defendant, the defendant can raise various affirmative defenses to liability, including the argument that the driver’s negligence contributed to the accident or to the harm that was caused. In most jurisdictions, a showing of driver contributory fault results in a reduction in the damages owed by the automotive defendant. In only a few jurisdictions, such a showing can result in the defendant being excused from liability entirely. Several scholars have envisioned retaining liability for product defects by adjusting the standard under which defectiveness is assessed. One of the most developed analysis is Mark Geistfeld’s proposal that the manufacturer would necessarily satisfy its tort obligation regarding the reasonably safe programming or design of the operating system if the aggregate, premarket testing data sufficiently show that the fleet of fully functioning autonomous vehicles performs at least twice as safe as conventional vehicles. To avoid liability for the crash of such a vehicle, the manufacturer must also adequately warn consumers about this inherent risk.[19] Once again, the risk involves systemized driving performance, and so aggregate driving data provide the appropriate measure and based on these data, auto insurers can establish the risk-adjusted annual premium for insuring the vehicle. By disclosing such a premium to consumers, the manufacturer would satisfy its obligation to warn about the inherent risk of a crash thereby eliminating this final source of manufacturer liability for crashes caused by a fully functioning autonomous vehicle.

In the very small percentage of accidents in which the losses are attributable solely or primarily to a defective vehicle, most state tort law usually state products liability law can be used to shift the losses from the driver and her first-party auto insurer to the vehicle’s manufacturer. If the vehicle manufacturer has no products liability insurance coverage for such losses but instead self-insures, most of those auto-accident costs will eventually be shifted to and spread over car buyers through increases in auto prices.[20] If, however, the vehicle manufacturer does have products liability insurance coverage for auto-defect claims, some of the auto-accident losses can then be shifted to the manufacturer’s insurer. Of course, over time, as auto products liability insurance premiums increase, those costs will be shifted back to automakers, who will again shift most of those costs back to auto consumers through higher auto prices and this shows the extent to which the liability potentially faced by manufacturers could have a substantial impact on the emerging market for automated driving technologies.[21]

However, some scholars reject the Product Defect standard and one of such scholars is David Vladeck who has worked through the rationale for such an approach in more detail and proposed what he refers to as “common enterprise liability”.  In Vladeck’s telling, this would consist of strict joint and several liabilities for all AV-related injuries, on the part of not only the manufacturer of the vehicle but also the makers of parts.[22] The technologically complex parts; the automated driving systems, the radar and laser sensors that guide them and the computers that make their decisions are prone to detectable failure, but those components may not be made by the manufacturer. Hence, from a cost-spreading standpoint, it is far from clear that the manufacturer should absorb the costs when parts and computer codes supplied by other companies may be the root cause.[23]


International liability standards with clear rules are needed. With concerns rising over the number and variety of state regulations, companies are increasingly looking to the federal governments for guidance. Representatives from Google, GM, Lyft, and Delphi testified before the US Congress on March 15 2016, urging Congress to pass a federal law concerning autonomous vehicles,[24] however, the passage of any federal legislation is unclear. There are multiple avenues for achieving uniformity. First, states could coordinate so that their regulations are consistent on the critical provisions that impact AV design and development. Alternatively, Federal Governments could pass regulations that preempt state laws. Commentators have suggested the need for federal preemption of state laws to create a uniform set of standards for AVs.[25]

International bodies also have a role to play in this juncture, in this light the European Union countries have a legal framework that will be well equipped to address and adapt to all the mentioned challenges in the legal regulation of autonomous vehicles that arise in the coming years, some legislative adjustments will probably be needed. However, having considered the massive reduction of injuries and fatalities caused by road accidents, and the other benefits of the autonomous technology, it is worth making those legal changes that will lead to clearer rules and practical reality. This in turn requires broad cooperation of lawmakers and technical professionals to achieve the most appropriate solutions. To address this problem, there should be enacted a strategy of promulgating nationally uniform safety regulations that function alongside the state tort and insurance systems. Perhaps this ironically can be largely solved by the technology itself for reasons that become apparent once we consider why the technology can cause an autonomous vehicle to crash before uniform regulations may be enacted.

 Any new approach will have to deal with the long and uneven transition to automated technology; impose substantial but appropriate financial responsibility for accidents on the manufacturers of highly automated vehicles, and provide satisfactory compensation to the victims of auto accidents in the new era. Carmakers at an industry show described autonomous vehicles as a future that won’t materialize unless legislators around the world create a new legal framework. To ensure that society reaps the maximum gains from the emerging AV market, it is paramount for governments to introduce new measures and regulations to manage the risks associated with AV.[26]


We are on the commencement of another new era, requiring another new legal regime. The new era of automated vehicles will eventually require a legal regime that properly fits the radically new world of auto accidents.[27] A liability system should not stifle innovation rather it should encourage it. Legal regulation of autonomous vehicles is a fairly complex object of research, all the more exciting, though. The most significant benefit of autonomous vehicles is a much safer driving environment. Accidents, however, will always be an aspect of motor vehicle travel and it must be decided who is to be held responsible in such cases. Stability coupled with a cost-spreading approach would doubtlessly serve that goal better than an uncertain fault – based liability system.

About the Author

Dele-Yaro Daniel Ademola ACArb is an associate arbitrator of the Nigerian Institute of Chartered Arbitrators and also a member of the international Law association. He is an excellent writer and astute researcher who has published articles on complex subject matters.


[1] M. A. Geistfeld, ‘A Roadmap for Autonomous Vehicles: State Tort Liability, Automobile Insurance, and Federal Safety Regulation’, <> accessed 14 July 2020.

[2] Adrian Ilka, ‘Legal Aspects of Autonomous Vehicles: An Overview’, <> accessed 13 July, 2020.

[3] Ibid.

[4] E. Hilgendorf, ‘Autonomous Cars and the Law’ <> accessed 15 July 2020.

[5] Christopher Mims, ‘Driverless Cars to Fuel Suburban Sprawl’ <> accessed 15 July 2020.

[6] Jerry Albright, ‘Marketplace of Change: Automobile Insurance In The Era of Autonomous Vehicles’ <> accessed 14 July 2020.

[7] M. Bertoncello and D. Wee, Ten ways autonomous driving could redefine the automotive world, (McKinsey & Company) 2015.

[8] National Highway Traffic Safety Administration, ‘Automated Driving Systems: A Vision for Safety’ <> accessed 7 January 2020.

[9] Ibid.

[10] Ibid.

[11] United States Department of Transportation, ‘Automated Vehicles for Safety’, <> accessed 30 October 2020.

[12] Tania Leiman, ‘Legal Liability for Accidents Involving Autonomous Vehicles’, < file:///C:/AI/2016_Leiman.pdf> accessed 15 July 2020.

[13] Ibid.

[14] A. D. Crane, A Survey of Legal Issues Arising from the Deployment of Autonomous and Connected Vehicles, [2017] (23) (2) Michigan Telecommunications and Technology Law Review.

[15] Reimann, Liability for Defective Products at the Beginning of the 21st Century, [2003] 51 (4) The American Journal of Comparative Law.

[16] John Villasenor, ‘Product Liability and Driverless Cars: Issues and Guiding Principles for Legislation’, <> accessed 18 October 2020.

[17] E. G. Marchant, The Coming Collision Between Autonomous Vehicles and the Liability System. [2012] 1321, 1333–35 Santa Clara Law Review.

[18] P. A. Garza, Wrinkles and Wrecks in the Age of Autonomous Vehicles, [2012] New English Law Review 581.

[19] Jerry Albright, ‘Market Place of Change: Automobile Insurance in The Era of Autonomous Vehicles ‘, <> accessed 20 November, 2020.

[20] S. K. Abraham, The Liability Century, (2008 Harvard University Press) 39.

[21] Ibid.

[22] D. C. Vladeck, Machines Without Principals: Rules and Artificial Intelligence, [2014] Vol. 89 (1) Washington Law Review 148.

[23] Ibid.

[24] Nathan Bomey, ‘Self-Driving Car Leaders Ask for National Laws’, <> accessed 3 December 2020.

[25] American Association of Motor Vehicle Administrators, ‘Autonomous Vehicles Best Practices Working Group’, <> accessed 3 December 2020.

[26] Ryan Narkashima, ‘Driverless Cars Need Legal Framework’, <> accessed 1 November 2020.

[27] S. K. Abraham, The Liability Century, (2008 Harvard University Press) 39-68.

Leave a Reply