This post is part of our special Call for Blog Posts “The Global Landscape of Digital Rights”. We welcome posts on national, international, and transnational debates about digital bills of rights, novel judgments, and other landmarks of the debates on digital rights around the world. Please send your contribution (1000–2000 words) to digitalconstitutionalist@gmail.com before 15 June 2022.
The so-called digital revolution adds, in social and democratic States under the rule of law, a double perspective regarding fundamental rights. On the one hand, the interaction between computing, electronics, telecommunications, mathematics, engineering, artificial intelligence (AI), robotics and other related sciences can create new and important powers for people, which can be part of the object of the fundamental rights already recognized. These new powers can be reflected in the laws that develop them or even may give rise to the need to promote constitutional changes to include some new rights, as has been the case of the ‘neurorights’, recently enshrined (October 25, 2021) in the Chilean Constitution. Its article 19.1 in fine provides that
scientific and technological development will be at the service of the people and will be carried out with respect for life and physical and mental integrity. The law will regulate the requirements, conditions and restrictions for its use in people, and must especially protect brain activity, as well as the information from it.
In this regard, the European Parliament
[c]alls on the Commission to consider an initiative on neurorights with the aim to guard the human brain against interference, manipulation and control by AI-powered neurotechnology; encourages the Commission to champion a neurorights agenda at the UN level in order to include neurorights in the Universal Declaration of Human Rights, concretely as regards the rights to identity, free will, mental privacy, equal access to brain augmentation advances and protection from algorithmic bias.
European Parliament Resolution on Artificial Intelligence in the Digital Age, paragraph 247.
At the national level, the Spanish Government has promoted the elaboration of the Charter of Digital Rights, July 14, 2021. However, it is a document without any legal value. As stated in the ‘Previous Considerations’, the Charter
does not try to create new fundamental rights but to outline the most relevant ones in the digital environment or describe instrumental or auxiliary rights of the former. The Charter is not normative, but its objective is to recognize the very new challenges of application and interpretation that the adaptation of rights to the digital environment poses, as well as to suggest principles and policies related to them in the aforementioned context. Thus, it also proposes a reference framework for the action of the public authorities so that, shared by all, allows us to navigate in the digital environment in which we find ourselves taking advantage of and developing all its potentialities and opportunities and conjuring up its risks. And contribute to the processes of reflection that are taking place at the European level and, with it, lead an essential process at a global level to guarantee a humanistic digitalization, which puts people at the center.
Nevertheless, specific mandates and prohibitions on the transparency of algorithms must be turned into genuine legal rules rather than mere ethical recommendations. This is the only way to guarantee clear and manageable access to the personal data as well as the right to share or transfer them in a simple manner; the establishment of effective mechanisms to combat discrimination and bias, with particular attention to the protection of vulnerable individuals and groups; the regulation of the use of AI by the police and judicial authorities as well as human seizures and surveillance through remote biometric identification systems; the liability and accountability legal regime; the intellectual property rights arising from the use of AI-related technologies and the rights of users and consumers, especially in the case of vital and advanced medical devices; the obligations of AI developers and deployers; the ways to appeal decisions made by AI systems, robotics or related technologies; the legal regime of AI in the field of defence and security; the regulation of the use of AI for different transport systems, in particular for the authorization of autonomous vehicles; the establishment of a national supervisory authority and so on.
In this regard, and as an example of a Charter with ‘legal force’, it is worth mentioning the approval, through Law No. 27/2021 of May 17, of the Portuguese Charter of Human Rights on the Digital Age. It includes, along with the classic rights such as the freedoms of expression, demonstration, association or participation in the digital world, other more recent rights, such as the right to be forgotten and the protection against abusive geolocation, the use of artificial intelligence and robots:
1. The use of artificial intelligence will be guided by respect for fundamental rights, guaranteeing a fair balance between the principles of explainability, security, transparency and responsibility, taking into account the circumstances of each specific case and establishing processes to avoid prejudice and discrimination.
Article 9 of the Portuguese Charter of Human Rights on the Digital Age.
2. Decisions taken by means of algorithms that have a significant impact on the recipients must be communicated to the interested parties, be subject to appeal and be auditable in the terms provided by law.
3. The principles of beneficence, non-maleficence, respect for human autonomy and justice, as well as the principles and values enshrined in Article 2 of the Treaty on European Union, namely non-discrimination and tolerance, shall apply to the creation and use of robots.
In any case, the national regulations of our geographical and political environment should be inserted in a genuine European legal framework which goes beyond ethical recommendations and indications. In fact, the precautionary principle, which guides the legislation of the European Union and must be central in any regulatory framework of artificial intelligence, plays a relevant role in this field. However, some argue that in this area—and in others of similar complexity, such as human embryonic research—the international regulation of AI should go beyond the traditional distinction between (binding) hard law and (non-binding) soft law.
In addition, computer science, AI and robotics and related digital technologies permeate three structural principles of the State and the fundamental rights related to them: the rule of law, the democratic and the social State. As far as the rule of law is concerned, the use of those sciences must be, in the first place, subject to the principle of legality, both in the broad sense of that term (submission to legal norms) and in the more precise one (regulation by-laws). Here we face different challenges, although the main one would be—we must say it again—subjecting to legal provisions, and not merely to ethical recommendations, the development, deployment and use of computing, AI, robotics and related technologies. In this regard, the European Parliament has been clear: ‘the risks currently posed by AI-based decision-making need to be addressed by the legislators’ and ‘calls for a regulatory environment for AI that provides effective governance and protection of fundamental rights’. However, ‘it is not always AI as a technology that should be regulated, but that the level of regulatory intervention should be proportionate to the type of individual and/or societal risk incurred by the use of an AI system; underlines, in this regard, the importance of distinguishing between ‘high-risk’ and ‘low-risk’ AI use cases; concludes that the former category needs strict additional legislative safeguards while ‘low-risk’ use cases may, in many cases, require transparency requirements for end users and consumers’.1Resolution on Artificial Intelligence in the Digital Age, paragraphs 16, 298, and 133.
Secondly, the judicial review of the administrative activity using information technology, AI, robotics and related technologies is a challenge that remains ongoing. What the Spanish Tax Agency intends to carry out could be a good example of it. According to its strategic plan 2020–2023,
it will take advantage of the potential offered by technology to complete the automation process in the processing of procedures initiated years ago. In this way, human resources will be freed up for the activities with the highest added value, a uniform treatment of taxpayers will be guaranteed, compliance with their obligations can be facilitated and contribute to eradicating tax fraud. For this, technologies such as natural language processing, advanced data processing and artificial intelligence will be key.
Decisions of independent administrative authorities—who should act as national supervisory authorities on the application of AI, according to the proposals of the European institutions—should also be subject to judicial review.
Thirdly, it is inherent to the rule of law the guarantee of some rights, initially of a “civil” nature but now also of a democratic and social nature. As we have seen in more detail in previous pages, there is little doubt about the need to ensure the enjoyment of such rights in a context dominated by the aforementioned technologies.
Concerning the democratic State, the pluralism of groups and associations, the respect for minorities and citizen participation in the different functions of the State (legislative, executive, jurisdictional)—participation that is both a right and a mandate of action for the public authorities as evidenced, for example, in Article 9.2 of the Spanish Constitution, in line with Article 3.2 of the Italian Constitution—are intrinsic to it. This being so, information technology, AI, robotics, and related technologies must not only respect but also contribute to the promotion of political and social pluralism by facilitating participation in electoral processes, virtual campaigns and online activism, as well as dialogue and debate between citizens and their representatives and governments. Likewise, measures must be adopted to guarantee a minimum of equality in electoral processes, transparency of campaigns, expenditure, messages and algorithms used in digital advertising in order to make political decision-making processes as fair and transparent as possible.
Finally, and concerning the social character of the State, the interventionist nature that characterizes it, as well as the search for the greatest possible real equality, must prevent the consolidation of biases. These can have broad effects and affect and discriminate many people without there being safeguard mechanisms such as those of social control that rule human behaviour.
With regard to citizens’ access to the digital world, its effectiveness will depend on the training actions and economic endowments provided by the public authorities. Thus, these kinds of rights could be conceptualized somehow as social rights since they demand performance from the State. In this line, the European Parliament resolution of 3 May 2022
calls on the Member States to make digital skills and literacy a component of basic education and lifelong learning; calls for a high-performing AI education system that fosters digital literacy, skills and digital resilience from an early stage, starting with primary education; emphasizes that the development of effective curricula for digital education requires political will, sufficient resources and scientific research; calls on the Commission to promote the introduction of AI and computational competence courses in all European schools, universities and educational institutions; highlights that such skills development is needed in adult education as much as in primary or secondary education; calls for a comprehensive and consistent policy initiative from the Commission and the Member States on AI skills and education at EU level, as well as for a legislative initiative on AI in the workplace.
Paragraph 205 of the European Parliament resolution of 3 May 2022.
In addition, the social dimension of rights, together with the democratic principle, gives a new perspective on the effectiveness of these rights. It should not only be vertical (against public authorities) but, at least in some cases, also horizontal (among individuals, within society). There must also be taken into account information technology, AI, robotics and related technologies and their impact, inter alia, on industrial relations and consumer law.
Suggested Citation
Miguel Ángel Presno Linera, ‘Fundamental rights and artificial intelligence in the social, democratic, and digital state under the Rule of Law’ (The Digital Constitutionalist, 23 May 2022). Available at <https://digi-con.org/fundamental-rights-and-artificial-intelligence-in-the-social-democratic-and-digital/>.
- 1Resolution on Artificial Intelligence in the Digital Age, paragraphs 16, 298, and 133.
