INTERNATIONAL JOURNAL OF TOURISM, TRAVEL AND HOSPITALITY LAW Let’s take now the case mentioned before of facial recognition at the airport, at security checkpoints. In such a case, as provided by the AIA, it is a biometric system, which is categorized as high risk, with the resulting liability regime. It may happen that the person undergoing the security check is mistaken for a terrorist and stuck at the airport. These are the well-known cases of mismatch. In such cases, the European Parliament Resolution of 202027 provided for a strict liability regime for high-risk AI systems. Today there is a change of course. On 28 September 2022, the European Commission published two proposals aimed at modernizing product liability rules in the digital age. The proposal is the first in the world specifically designed to address compensation for damage caused by artificial intelligence (AI) systems. Two Directives are proposed: one on adapting non-contractual civil liability rules to AI (the AI liability Directive); and the other for defective products (the Product Liability Directive). The proposed Directives are said to complement each other and the EU AI Act. The Product Liability Directive addresses a producer’s “no fault” or “strict” liability for defective products (including AI) and associated compensation for damages. Whereas the AI Liability Directive is concerned with liability arising from “wrongful behavior” by AI systems28. The line between the two responsibilities turns out to be very thin and unclear. However, if anything certain can be said, it is that there is no longer a clear differentiation of the liability regime applicable to high-risk and low-risk AI systems. This difference, on the other hand, was apparent in the previous recommendation. However, the Proposal for an AI Civil Liability Directive of September 28, 202229 provides for fault-based liability for both high-risk and low-risk systems, albeit with some differentiations (which apparently sounds strange enough). In conclusion, in the light of the examples given before, there is something wrong. There is no harmony between the Proposals on artificial intelligence and 27 European Parliament resolution of 20 October 2020 with recommendations to the Commission on a civil liability regime for artificial intelligence (2020/2014(INL)), available at the link: https://www.europarl.europa.eu/ doceo/document/TA-9-2020-0276_EN.html#title1. 28 In favor of a “strict liability” regime for AI systems see Vladeck, D.C. (2014) ‘Machines without Principals: Liability Rules and Artificial Intelligence’, Wash. L. Review, 117/2014, pp. 117-150. 29 Proposal for a Directive of the European Parliament and of the Council on adapting non-contractual civil liability rules to artificial intelligence (AI Liability Directive). The text is available at the link: https://ec.europa.eu/ info/sites/default/files/1_1_197605_prop_dir_ai_en.pdf.
RkJQdWJsaXNoZXIy MTE4NzM5Nw==