The European Parliament is about to adopt a position on the very popular issue of autonomous transport on Tuesday 15 January. While the high-level working group on ethical issues is very evasive on the issue of liability, the draft report - which should be adopted as it stands - is in favour of implementing manufacturer's liability.
The non-legislative draft report, which covers all modes of transport and public transport, pays particular attention to connected cars (see EUROPE 12073). It acknowledges that "passenger cars with superior automated driving functions" will most certainly be on the European road network from 2020 onwards and "that several countries around the world [including the United States, Australia, Japan, Korea and China] are in the process of marketing connected mobility and automated mobility".
An "objective" responsibility
Despite the very sensitive nature of this issue, the draft report that was already adopted by the Transport Committee on 22 November (33 votes in favour, 3 against) attempts to answer the difficult question of liability in the event of an accident caused by a totally autonomous vehicle.
Two paragraphs are particularly interesting: the first recommends that consideration be given to whether a transfer of responsibility should not be made, bearing in mind that only a very small proportion of accidents today are attributable to technical causes. This would imply, the report notes, the introduction of "manufacturer's liability linked simply to the risk arising from the introduction of an autonomous vehicle as an element of objective endangerment" and, therefore, to examine whether specific road safety obligations imposed on the vehicle owner as well as obligations regarding instructions given to individual drivers can duly compensate for this transfer of liability.
The second paragraph goes one step further, noting that a strict liability insurance scheme (called "no-fault" because it is based on cause) would make it possible, in the event of damage caused by an autonomous car, to provide a response to current legal loopholes and gaps.
Reflections and research
It should be recalled that the thorny issue of responsibility and, more generally, the ethical dimension of artificial intelligence have been the subject of a draft report of the High Level Group of Experts on Artificial Intelligence (see EUROPE 12162). However, the 52 contributors to this report are relatively evasive on the notion of accountability, postponing this issue until mid-2019, when the group of experts will issue recommendations for AI policy and investment.
Also noteworthy is the very first meeting, on 10 January in Barcelona, Spain, of the AI4EU project to create a focal point for artificial intelligence (AI) resources, including data repositories, computing power, tools and algorithms. This project, led by the French company Thales, brings together 79 leading research institutes, SMEs and large companies in 21 countries. (Original version in French by Sophie Petitjean)