One and a half year after presenting its first-of-a-kind Proposal for a Regulation of the European Parliament and of the Council Laying Down Harmonised Rules on Artificial Intelligence (Artificial Intelligence Act) and Amending Certain Union Legislative Acts (EU AI Act draft), the European Union Commission (hereinafter “EU Commission”) introduced on September 28, 2022, its Proposal for a Directive of the European Parliament and of the Council on adapting non-contractual civil liability rules to artificial intelligence (the Directive proposal). This proposed Directive aims to harmonize the national liability rules for artificial intelligence (AI) and make it easier for victims of AI-related damage (natural persons, businesses and any other public or private entities) to get compensation. This will ensure that they are no less protected than victims of traditional technologies. Just like the EU AI Act, which will have an extra-territorial reach, the liability regime provided for in the draft Directive, as introduced into the individual national law corpus of the EU Member States, will be applicable to EU and non EU-based companies alike when then conduct business in the EU.
As a reminder, a directive is a legislative act that sets out a goal that all EU countries must achieve and it is up to the individual EU Member States to include the norms through their national laws. In the case of AI, the goal of the Directive proposal is to provide the flexibility “to enable Member States to embed the harmonised measures without friction into their national liability regimes 1 ”. It strives towards helping private law to take the much-needed turn allowing it to meet the needs of the current transition to a digital economy 2 . Of course, private enterprises and businesses around the world are main actors in this shift.
What is the difference between the Directive proposal and the EU AI Act proposal?
Whereas the EU AI Act proposal seeks to reduce risks for safety and protect fundamental rights, the Directive lays down rules for individuals to sue companies for damages resulting from an AI system.
As the EU Commission underlined “[s]afety and liability are two sides of the same coin: they apply at different moments and reinforce each other 3 ”. Indeed, guaranteeing effective compensation contributes to the protection of the right to an effective remedy and a fair trial 4 . Thus, the Directive proposal would “contribut[e] to the enforcement of the requirements for high-risk AI systems imposed by the AI Act 5 ” and would be adopted as a measure to ensure the establishment and functioning of the EU internal market 6 .
The two proposed legal texts will thus work together. In addition, once adopted, the Directive will ensure that operators and users of AI systems in the EU wouldn’t be faced with 27 different liability regimes. If it were the case, it would lead to different levels of protection and allow for a distorted competition among businesses from different EU Member States 7 .
The EU Commission underlined the need to adopt specific legal norms for AI liability considering the inadequacy of the current national liability rules for handling liability claims for damage caused by AI-enabled products and services. The unique characteristics of AI systems, which include “complexity, autonomy and opacity […] may make it difficult or prohibitively expensive for victims to identify the liable person and prove the requirements for a successful liability claim 8 ”. Thus, the Directive would alleviate the burden of proof in relation to damages caused by AI systems and establish broader protection for victims, be it individuals or businesses 9 .
What does the Directive proposal provide for?
The EU AI Liability Directive follows the work laid out in the EU AI Act to ensure consistency. More specifically, art. 2 (6) (b) of the Directive provides that claims for damages can be brought forth not only by the injured person but also by persons that have succeeded in or have been subrogated into the injured person’s rights, such as an insurance company or the heirs of a deceased victim.
Furthermore, article 2(6)(c) sets forth that an action for damages can also be brought by someone acting on behalf of one or more injured parties, in accordance with EU or the national law of an EU Member State.
The Directive also introduces a specific mechanism for the disclosure of evidence in order to “to provide persons seeking compensation for damage caused by high-risk AI systems with effective means to identify potentially liable persons and relevant evidence for a claim 10 ”. This aims to help victims access relevant evidence. Indeed, article 3 (1) of the Directive provides that
“Member States shall ensure that national courts are empowered, either upon the request of a potential claimant who has previously asked a provider, a person subject to the obligations of a provider pursuant to [Article 24 or Article 28(1) of the AI Act] or a user to disclose relevant evidence at its disposal about a specific high-risk AI system that is suspected of having caused damage, but was refused, or a claimant, to order the disclosure of such evidence from those persons.”
In addition, the EU AI Liability Directive introduces a rebuttable “presumption of causality”. According to this principle, the court is to presume that the non-compliance with a certain obligation relevant to the harm caused by the AI system, if the causal link is likely, caused the damage.
When will these legal obligations become applicable?
It is likely that it will take years for the proposal to become law, after potentially undergoing important changes to its text, as it still needs to undergo the EU legislative process. However, along with the EU AI Act draft (which is set to become law in 2023 or 2024), the Directive demonstrates a clear momentum in the legal world for the adoption of severe norms for AI, in the EU and abroad.
Accordingly, organizations must keep abreast of the major legal changes that are coming in the AI sector and keep in mind that they will have far-reaching ramification, as AI systems are likely to be found in most spheres.
Moreover, the adoption of EU legislation on AI will likely have a ripple effect around the world, which means that most companies will have to adapt their activities to stringent new obligations, globally.
At the Law Office of S. Grynwajc we help companies navigate through transatlantic legal and regulatory landscapes. Our lawyers are admitted to practice in both the EU and the US and are therefore ideally placed to help you operationalize new EU rules into your day to day activities. If you need assistance understanding what these new rules mean for you and how to prepare for them please contact us. We’d love to help.
1 Directive of the European Parliament and of the Council on adapting non-contractual civil liability rules to artificial intelligence (hereinafter “the Directive proposal”), p. 7.
2 Directive proposal, art. 1.
3 Directive proposal, p. 2.
4 EU Charter of Fundamental Rights, art. 47.
5 Directive proposal, p. 3.
6 Treaty on the Functioning of the European Union, art. 114.
7 Directive proposal, p. 6.
8 Directive proposal, p. 1.
9 Directive proposal, p. 6.
10 Directive, introduction, point 3.