Autonomous Intelligent Cars on the European intersection of liability and privacy
Download 132.72 Kb.
|
- Navigate this page:
- Autonomous Intelligent Cars on the EU roads
| DRAFT 23 March 2015, Roeland de Bruin.1
Autonomous Intelligent Cars on the European intersection of liability and privacy. Consumer cars are increasingly being equipped with technology automating certain aspects of driving. Examples of such technologies include lane keep assistance, emergency braking, parking assistance and adaptive cruise control. In the near future, higher levels of car-automation will become available, eventually leading to the introduction of autonomous intelligent cars (AIC). AIC could amongst other things contribute to safety on the roads, and decrease the number of car accidents (93% of traffic accidents are caused by human failure, 2 leading to 1.3 million deaths and 50 million serious injuries worldwide per year).3 A definition of AIC consists of three elements. Autonomy relates to the level of human intervention necessary for operation, which can be seen as a spectrum: a lower need for human intervention implicates a higher level of autonomy. Intelligence relates to the ways in which a system can perceive its surroundings, and is able to adapt behaviour to changing environments. It includes the ability to learn, to reason and to solve problems.4 Ultimately, AIC may become capable of perceiving data and information, which are able to learn and make decisions based upon their experience, used for transportation of goods or people to a certain destination without the need of human intervention. Although it is likely that specific aspects of car-technology will become more automated in the near future, many technological challenges must be met to make fully autonomous intelligent driving available on a large scale to consumers. Besides technological challenges, the introduction of (gradual) autonomy in automotive technology also implicates a number of ethical, legal, societal and economic issues. The legal challenges include many fields law. The question can be raised if, and to what extent the forthcoming introduction of Autonomous Intelligent Cars, given the current regulatory framework in the EU, would lead to desirable outcomes, in relation to the core values underpinning this regulatory framework. These core values include the fair apportionment of risks inherent to modern technology, the protection of consumers and the stimulation of innovation in terms of economic growth. Furthermore, it is the question in which ways and on what levels the regulatory framework can be optimized, to equip both the development of AIC-technology and accepted introduction thereof in society. In this respect, it must be taken into account that legislators tend to respond to the introduction of new, disruptive technologies such as photography, the VCR and software,5 with the instalment of case specific legislation, without the level of flexibility necessary for anticipating new innovations as well as the societal adoption thereof.6 This paper will discuss the possible impact of the current regulatory framework on the development of AIC and deployment thereof in society by looking at the intersection of civil liability law and data protection law in the European Union based on a – simple – case study. The question will be addressed in which ways the current EU regulatory framework on establishing liability for damages caused by AIC (insofar that is present to date), using for example vehicle tracing technology, may interfere with the right to data protection of consumers of AIC-technology. An analysis will be made of the harmonized rules on product liability law, followed by a case study of (non-harmonized) liability for car owners/possessors in the Netherlands. This will show that AIC-technology in itself will contribute to answering the question as to how to establish liability. For instance AIC equipped with black boxes,7 or vehicle tracing technology recording (and communicating) sensory information, decisions of the systems and input from the driver will be used to establish the cause of an action leading to damage. The fact that AIC technology can also assist in allocating liability may form an incentive, for instance, to insurance companies to insure certain risks involved in AIC, which may in turn contribute to development and deployment of the technology. However, the same fact that AIC will be able to process and store massive amounts of (personal) data, possibly being connected to other AIC’s, or (interconnected) networks, forms a potential threat to the privacy of its drivers, which may hinder acceptance of the technology. Finally, this paper will conclude with some remarks on potential challenges for the EU regulatory framework given the core values of the allocation of risks, consumer protection and the EU Horizon 2020 ambitions on innovation and economic welfare, in view of the pending introduction of Autonomous Intelligent Cars. Table of Contents 1.Introduction 1 2.Autonomous Intelligent Cars on the EU roads? 4 3.Civil liability for damages caused by Autonomous Intelligent Cars 6 3.1A Scenario 6 3.2Harmonized framework on Product Liability 6 3.2.1Regulatory framework: Core Values of the Product Liability Directive 6 3.2.2Addressing the questions under the Product Liability Directive 9 3.3Less harmonized: framework on liability for motor vehicles 10 3.3.1France 10 3.3.2Netherlands 11 3.3.3United Kingdom 12 3.4Interim observations 13 4.AIC tracing technology and information privacy 15 4.1AIC tracing technology? 15 4.2Current framework 15 4.2.1EU Data Protection Directive 16 4.2.2Forthcoming EU General Data Protection Regulation 19 4.3Interim observations 20 5.Summary and conclusions 21
The European Union aims to promote innovation, which is seen as a key driver of economic welfare and growth.8 It is observed that, while the EU market is the largest in the world, it is not innovation-friendly enough. The EU Member States are together spending 0.8% of GDP less than the US, and 1.5% less than Japan yearly in research and development.9 A number of measures taken to stimulate innovation is set out in the Digital Agenda for Europe.10 Investing in the development of robotics in Europe lists high on the Digital Agenda and it aims to contribute to inter alia productivity and competitiveness, reindustrialisation, health, environment and security.11 Autonomous Intelligent Cars form an example of the development of robotics in Europe. Besides furthermore contributing to safety on the European roads, AIC may be beneficial for efficient use of the road network, may lead to reducing CO2-emissions and could contribute to mobility of disabled people.12 Not everyone is optimistic about a driverless future. It is stated that while in general AIC will be beneficial to road safety, other risks will follow from the introduction of autonomous vehicles, which will (when interconnected) for example be vulnerable to hacking. Also businesses and jobs that rely on the way in which logistics processes for persons and goods are organised, such as taxi and transportation markets will change significantly and eventually become obsolete after the autonomisation of driving.13 It is also observed that accident risks will increase when autonomous and non-autonomous cars co-exist on the same roads. Furthermore, some are just reluctant to give up the pleasure of driving.14 Many steps towards automating driving have already been made. Modern cars are often equipped with some of the following technologies: lane-keep-assistance, electronic blind spot assistance, traffic jam and queuing assistance, adaptive cruise control, emergency breaking and crash avoidance. There are prototypes available which are able to drive without someone controlling the car. Google is currently pioneering self-driving car technology, and has announced to put a fully functioning prototype of an autonomous intelligent car to road tests (Bay Area, California, United States) in the beginning of 2015.15 Also in the European Union, car manufacturers concentrate on the development of AIC-technology.16 For example Scania is testing “Platooning” technology: on 9 February 2015 a road train of self-driving trucks was deployed on the Dutch roads, which were autonomously following a human controlled truck heading the convoy.17 Volvo has planned to deploy 100 cars which are claimed to be able to take over all aspects of driving in Sweden by 2017.18 In Germany, a part of the A9-Autobahn between Munich and Berlin is reserved for extensive testing of autonomous vehicles in the coming years.19
Fully autonomous vehicles are not available to the public yet, and it is predicted that it will take at least some years if not decades before AIC will be commonplace on the European Roads. The development of autonomy in cars is observed to be a spectrum. Starting from ‘driver only’, the next steps toward ‘full autonomy’, a stage in which drivers eventually become redundant, may be ‘assisted driving’, ‘partial autonomy’ and ‘high autonomy’.20
|
