AI evidence and the future of motor vehicle accident disputes

Abstract

At first glance, UK legislation regulating automated vehicle (AV) claims appear to provide a straightforward mechanism via an insurer for any road user to make a claim for injury or damage caused by an AV whilst engaged in self-driving mode. However, this paper will demonstrate how in the event of the insurer denying a claim or the manufacturer/insurer alleging that the Claimant contributed to the accident, those seeking compensation will have no alternative but to take the expensive and time-consuming action of pursuing the matter through the courts. 

 The UK has not (at the time of writing) amended its product liability legislation to include AI and software. Consequently, if there is a vehicle fault relating to the AI, Claimants must pursue the claim in court via a negligence claim.  Relevant to this, the UK has also introduced legislation which creates a benchmark of safety for AVs which is that AVs should drive to the standard of a competent and careful human driver.  An AV being involved in a collision while driving itself does not mean the AV has fallen below the safety standard. 

Therefore, if an AV operator or road user alleges that the AV’s performance fell below the required safety standard, to prove this, the Claimant would require access to vehicle data and expertise to interpret the data.  Parties not in possession of the AV data necessary to prove a claim will have to seek access to data through the manufacturer or other AI designer/provider. Where this is not voluntarily forthcoming, production of this evidence must also be pursued through the Courts. While there are laws regarding compulsory incident recording data in AVs, the parameters are narrow, and this data may not reveal how an incident occurred. Where the claim alleges faulty AI, which is denied by an insurer, the existing legal presumptions about the reliability of computers, combined with the inexplicable nature of deep learning algorithms, presents an almighty challenge for both a Claimant to present a case and for the Court to understand the evidence before it.  This paper uses a learning scenario to explore how courts may deal with AV accident litigations, the types of challenges which may arise and the impact this may have on AV users and other road users who make a claim following a collision where a vehicle was driving itself.