< Back

Share |

Driverless cars and product liability

Driverless cars are among the most significant advancements in technology set to bring about changes to our existing product liability regimes.

November 2017

Driverless cars and, more currently, semi-autonomous vehicles already being tested and used on the roads raise new questions regarding product liability. Such questions challenge the traditional liability routes for losses arising out of defective and even unsafe products under our existing product liability laws.

Consumers can expect to see driverless cars (capable of fully autonomous operation) on the roads in the next decade. While improving road safety is the ultimate goal, manufacturers must accept the increased likelihood of damage and/or injury being caused by the technology used in these vehicles (without driver involvement). Manufacturers and technology providers therefore need to start addressing the importance of understanding the risks and liability issues that these technological advancements present as they continue to evolve.

When things go wrong and defects are discovered resulting in loss and damage, where will the product liability landscape lie? Will human error be completely ruled out or does the driver have a responsibility to maintain and service the vehicle in compliance with the manufacturer's instructions? We will likely be moving away from the traditional user/driver error to defaults caused by defects in the technology 'products' used in the vehicles. As a result, liability will shift to the manufacturer and the spotlight will be firmly placed on all of those parties in the supply chain who play a part in the development of the automated technology.

Insurance companies are also intended to be primarily liable for accidents caused by automated vehicles under the recently published Automated and Electric Vehicles Bill (AEV) by Parliament. If the automated vehicle is insured, then insurance companies will be liable for death, personal injury or damage to certain property suffered by an inured party arising from accidents whether the vehicle is being driven by the driver or in driverless mode. Insurers will be able to exclude or limit liability where insureds make unauthorised software alterations or where they fail to install safety critical updates where they ought reasonably to have known they were safety critical. In addition, insurers will have a right to raise a claim against those responsible for an accident in relation to any money they have paid out in damages to those affected by the accident. Insurers will look to pass on that liability and recover sums from insurance claims from automotive manufacturers under existing product liability laws. This is, therefore, a key area of development in the law of product liability (read more on insurance and liability for CAVs in our article).

Product liability – the traditional route

Up until now it has been fairly clear where a claim for product liability might exist in the context of automotives. Certain parts of a car are driver-operated and malfunction can be a result of consumer misuse leading to a potential personal claim against the driver (in personal injury, negligence or criminal action). Liability rests with the driver rather than the manufacturer who relies on their relevant insurance policy where applicable. Faulty parts as a result of a manufacturing design failure, defect or a failure to use reasonable care and skill, however, lead to a claim against the manufacturer under product liability laws.

Under the Consumer Protection Act 1987, (CPA) a manufacturer is strictly liable in respect of defective products and the injured party does not have to establish fault; only that a defective product caused the loss. A product is defective under the CPA if it is not as safe as persons are generally entitled to expect taking into account the purpose for which the product has been marketed, any instructions for use or warnings, what might reasonably be expected to be done with the product at the time when the product was supplied – a product is not unsafe simply because a safer product was later developed or industry safety standards were raised after it was supplied.

The test is, therefore, one of consumer expectation and this will be more complex in the context of autonomous vehicles where users are unlikely to have any significant understanding of the technology products used. As to their expectations on safety, these could be higher or unrealistic. Manufacturers will need to ensure that they inform consumers sufficiently as to how the automated features should be used safely and explain any potential risks. Manufacturers and software providers should consider taking advice on the formation of product instructions and warnings given to consumers with the automated vehicles (particularly in the user manuals) as such documents will become increasingly important in the context of product liability claims.

Autonomous vehicles – setting a new scene for product liability?

Automation in cars continues to develop at a rapid pace. Consumers have been, for some time, enjoying elements of automation with cruise control functions, parking assist technology, on-board cameras and automated braking systems. These combine human input with technology assistance and allow the driver to take control of the vehicle if necessary. The goal for automated vehicles is to ultimately improve safety and reduce the number of accidents on the road caused primarily by driver errors. That being so, it follows that software developers and technology providers of automated components will be firmly placed in the spotlight for consumer safety. We can expect to see questions of liability being raised with a number of parties being responsible for components of the automation, and liability shifting from user error to supplier default.

Under the product liability route, how will liability be determined between the driver, manufacturer and technology/software providers? There could be multiple parties liable. At the moment, car manufacturers (such as Volvo) have spoken out to say they will accept liability for losses arising out of faults in the design/functionality of the automated technology assistance. However, how will this work in practice – can such assurances be legally enforced through warranties/indemnities, third party/contribution claims? What protections will need to be in place for the consumers/insurers?

Evidence will be a new issue - how will liability be proven, what evidence will be available? The 'black box' of data will be key. Expert evidence will also be needed to opine on any limitations in the software and how faults could have occurred. What will be made of software updates and will this play into the State of the Art/development risks defence under the CPA absolving manufacturers from liability for products which cannot be said to be defective if the alleged defect was unknown in the industry at the time? - " The state of scientific knowledge at the relevant time was not such that a producer of products of the same description as the product in question might be expected to have discovered the defect if it had existed in his products." Can users still be held responsible if they have failed to service the vehicle and/or maintained it in accordance with manufacturer instructions? The AEV Bill suggests they can.

Time will tell how these issues will be considered in the legislative framework surrounding driverless cars and we continue to monitor them with interest. In the meantime, there are some practical steps which manufacturers can consider now to help in protecting their position on liability. These include:

  • reviewing existing testing procedures and risk assessments to ensure they adequately deal with new risks arising out of the shifts in liability from driver to manufacturer;
  • keeping documentation up to date to justify steps taken (in respect of the design and manufacturing process and testing phase) and prove the manufacturer is acting reasonably with regard to safety and in line with the state of the art or any relevant industry standards;
  • reviewing policies for responding to product risks – do they adequately deal with urgent product recalls for automated technology or hacking incidents? Is active monitoring of security and safety issues arising from connected devices required? Can all components of the software assistance be traced to the correct manufacturers/suppliers? How will all relevant parties in the supply chain of an automated vehicle be informed of any potential safety issue?; and
  • reviewing product instructions and warnings provided to consumers and making any necessary amendments to ensure that consumer expectations on safety are adequately covered.

These are all issues that manufacturers and suppliers in the automated vehicles supply chain should be considering now to keep up with this rapidly developing technology and changing product liability landscape.

What's next?

The AEV Bill should make its way to enactment (although possibly with amendments). Insurance companies see driverless cars as the biggest change to the insurance automotive sector and policies have already been developed for drivers. The UK government research indicates that the UK market for automated vehicles will be worth £28bn by 2035.

An EC consultation is also underway to review the existing EU product liability legislation and how this can be amended to ensure it is fit for purpose to deal with claims arising out of technological advancements including connected cars, connected medical devices, 3D printing, internet of things etc. The issues arising out of this go beyond the scope of this article but include whether the existing Product Liability Directive (which the UK adopted under the CPA) adequately covers the cross-over of these developing technologies with product liability (e.g. do certain definitions need to be reconsidered such as whether "software" is a "product" or a "service" and is "defect" sufficient to cover all the types of components used, particularly with devices connected to the internet?). That said, whether the UK is even bound by any changes to the EU product liability regime remains to be seen post Brexit. Our view, however, is that the UK will adopt any changes to these laws to ensure its chances of becoming one of the leading jurisdictions in the manufacture and supply of automated vehicles and maintaining consumer confidence across the EU.

Manufacturers are only now starting to become aware of the need to address the new liability issues and a balance will need to be struck between protecting the consumer without impacting on innovation in this exciting and rapidly evolving industry.

If you have any questions on this article please contact us.

connect car
Katie Chandler

Katie looks at the challenges to the traditional liability routes for losses arising out of defective products in connected and autonomous vehicles.

"We can expect to see questions of liability being raised with a number of parties being responsible for components of the automation, and liability shifting from user error to supplier default."