The Council of Europe’s Artificial Intelligence Convention: Implications for Health and Patients

The Council of Europe, the most important international human rights organization on the European continent, currently is drafting a Convention on Artificial Intelligence, Human Rights, Democracy and the Rule of Law (AI Convention). The Convention aims to protect fundamental rights against the harms of Artificial Intelligence (AI), and is expected to become a global leading convention, as non-European states such as the United States (U.S.) are considering becoming signatories.

As health care is one of the top industries for AI, the forthcoming AI Convention will have important implications for the protection of health and patients. This post gives a brief outline of the background, scope, and purpose of the AI Convention. It goes on to flag common human rights issues associated with medical AI and then touches upon the most important health rights implications of the current text of the AI Convention.

  1. Background, Scope, and Purpose

Since 1949, the Council of Europe and its human rights court have played an important role in the protection and promotion of human rights in Europe through a long list of treaties, resolutions, and court decisions (some of which relate to new technologies). The Council’s scope is much broader than the European Union’s (EU): it has 47 member states (including all 27 EU Member States) and 5 observer states (including the U.S. and Japan). The observer states are not bound by the Council of Europe but can choose to sign on to the AI Convention — committing to its rules as well.

The AI Convention, which is expected by the end of 2023, will exist in parallel to and aims to complement the EU’s forthcoming AI Act and other regulatory initiatives, including the U.S. AI Bill of Rights. The AI Act focuses on the marketing of products using AI within the EU internal market, while the AI Convention focuses on the protection of the fundamental rights of people affected by AI systems. The Convention is principle-based and will — unlike the AI Act and the AI Bill of Rights  — introduce legally binding individual human rights, which also will apply to citizens of non-EU states. The exact scope of the Convention, however, is still under discussion, with states like the U.S., the U.K., and Japan advocating that it remains limited to the public sector.

  1. Medical AI: Challenges for Health and Human Rights

While medical AI can improve the quality and efficiency of health care, it also comes with several risks to health and human rights. System errors in widely used AI could cause mass patient injuries, for example, in AI-powered robot-assisted surgery. Unrepresentative or low-quality datasets used to train AI models may lead to incorrect medical decisions causing health damage, such as false diagnoses or wrong treatment decisions. When datasets reflect existing biases in society, specific populations face greater risk of health harms, reinforcing discrimination and health inequality. The enormous ‘data hunger’ of medical AI may also affect medical privacy, and the opaque nature of many AI applications may put existing health practices and other patients’ rights under pressure, such as the provision of information, informed consent, and legal redress.

  1. The Meaning of the AI Convention for Health and Patients

Lifecycle Regulation of General AI Systems

The definition of the proposed AI Convention is rather broad and includes most medical AI applications: “a machine-based system that is capable of informing or autonomously generating outputs such as predictions, recommendations or decisions using machine and/or human-based data and inputs and influencing the environment that it interacts with.” This means it covers applications like ChatGPT-assisted diagnosis, AI-based fall detection in elderly care facilities, AI-assisted surgery, and menstruation cycle trackers.

If the scope would be limited to uses in the public sector, it would still cover a fair amount of medical AI applications, such as automated disease surveillance, social benefits distribution, and most COVID-19 apps, plus many public health care systems. Moreover, the AI Convention applies to the entire lifecycle of AI: the design, development, and application of AI systems. This is especially important for adaptive AI systems that evolve over time.

This sets the Convention apart from the EU’s AI Act, which focuses on regulating the development phase, and only stipulates strict rules for “high-risk” AI medical devices.

Enforceable Rights for Patients

The AI Convention departs from the understanding that all human rights are potentially impacted by the growing use of AI systems, instead focusing on the specific rights to non-discrimination, privacy and data protection, transparency rights, and the right to redress for harm. It establishes these as enforceable rights — filling an often-criticized gap of other AI regulations.

Foremost, the Convention stipulates that all AI systems undergo a human rights assessment before deployment. Then, the Convention recognizes the right to non-discrimination, which is important for health equity — i.e., in cases of limited internet access in remote areas and insufficient digital skills to understand AI tools. The Convention protects privacy and data protection rights, setting high requirements for data protection, which is important for patients who often share intimate data with their doctor. The Convention also equips patients with the right to redress for harm, requiring robust frameworks for accountability and liability — a much-needed and currently absent solution for medical AI causing harm. The transparency rights put forward by the Convention contribute to the creation of a relationship of trust between patient and doctor and ensure informed consent to medical treatment.

Especially for health care — a sector that is mostly regulated locally — minimal human rights standards are important, as they give patients in all Member States the means to arm themselves against potentially harmful medical AI.

The Council of Europe Cares About Health

Generally, the AI Convention is promising for patients because of the Council of Europe’s long track record of protecting the health and human rights of patients. The Council has published many recommendations on medical-ethical issues aiming for stronger human rights protections in the medical field (i.e., health governance, organ donation, medical data protection). The Human Rights Court has enforced protection for patients against — inter alia — forced medical treatments, medical data breaches, and health discrimination. It has also established strong obligations for states to actively protect the health and patients’ rights of their citizens. Many states have changed their health practices and laws following Court orders, for example, for involuntary psychiatric commitment.

In short: health and patients’ rights benefit from a human-rights approach. The Council of Europe’s AI Convention, being the first binding fundamental rights instrument for AI negotiated at such a widespread scale, holds the potential of uplifting health and patients’ rights worldwide, when medical AI becomes common practice in health care.

Originally posted on Bill of Health, the blog of the Petrie-Flom Center at Harvard Law School. Please cite as:

Hannah van Kolfschooten, ‘The Council of Europe’s Artificial Intelligence Convention: Implications for Health and Patients’, 18 April 2023, Bill of Health, <;.

EU Regulation of Artificial Intelligence: Challenges for Patients’ Rights

In order to create a well-functioning internal market for Artificial Intelligence (AI) systems, the European Commission recently proposed the Artificial Intelligence Act. However, this legislative proposal pays limited attention to the health-specific risks the use of AI poses to patients’ rights. This article outlines that fundamental rights impacts associated with AI such as discrimination, diminished privacy and opaque decision-making are exacerbated in the context of health and may threaten the protection of foundational values and core patients’ rights. However, while the EU is facilitating and promoting the use and availability of AI in the health sector in Europe via the Digital Single Market, it is unclear whether it can provide the concomitant patients’ rights protection. This article theorizes the Europeanization of health AI by exploring legal challenges through a patients’ rights lens in order to determine if the European regulatory approach for AI provides for sufficient protection to patients’ rights.

In 2021, the European Commission proposed the Artificial Intelligence Act. In her recent article for Common Market Law Review, Hannah van Kolfschooten expresses concerns about this legislative proposal with regard to patients’ rights. The aim of the proposed act is to create a well-functioning internal market for Artificial Intelligence (AI) systems. However, Van Kolfschooten argues that limited attention is paid to the health-specific risks the use of AI poses to patients’ rights. She states that fundamental rights violations associated with AI are exacerbated in the context of health and may threaten the protection of patients’ rights. While the EU is facilitating and promoting the use and availability of AI in the European health sectors, it remains unclear whether it is simultaneously able to protect patients’ rights against the risks that the use of AI poses.

Please cite as: Hannah Van Kolfschooten, ‘EU regulation of artificial intelligence: Challenges for patients’ rights’, (2022), 59, Common Market Law Review, Issue 1, pp. 81-112.

How the Parliament’s AIDA Report Misdiagnoses the Risks and Benefits of Health AI

In November 2021, the European Parliament’s Special Committee on AI in a Digital Age (AIDA Committee) put forward its draft report on artificial intelligence (AI) in a digital age. The report advocates for a very permissive approach to the regulation of AI to stimulate innovation and foster the competitiveness of the EU. With this, however, it understates the possible risks specific to the development and use of AI in the context of health and lacks actual solutions to translate potential into action. We identified five shortcomings and hereby appeal to the regulator to address them.

Jorge Félix Cardoso, M.D., M.A. is a Parliamentary Assistant at the European Parliament

Hannah van Kolfschooten, LL.M. is a PhD Researcher at Law Centre for Health and Life, University of Amsterdam.

Diogo Nogueira Leite, M.Sc. is a PhD Researcher in Health Data Science at the Center for Health Technology and Services Research, University of Porto.

Tjaša Petročnik is a PhD Researcher at TILT/TILEC, Tilburg University

Henrique Vasconcelos, M.D.

The authors write in their personal capacity.

Health is broader than healthcare

When talking about AI in or for health, we are not only talking about a cancer diagnosis or the personalisation of therapeutics in the clinic; the scope of health provision is expanding from formal healthcare systems into our smart devices, actively involving (pre-)patients and consumers. Think about increasingly AI-driven fitness apps, symptom checkers, disease management tools and so on, which have the potential to widen access to health-enhancing resources by cutting conventional healthcare ‘gatekeepers’ out of the equation.

These tools are, however, in the report seen as mere means to relieve pressure on healthcare systems, even though they might affect the health attitudes and behaviours of users or even result in serious harm if not performing optimally. Additionally, as health data is scarce and desirable, we can expect such consumer-facing AI tools to progressively act as an avenue to obtain such data. Misuse might result in privacy violations, civil rights discrimination based on health records, or health inequality increases. AI regulation should thus take into account that AI-based health practices are also taking place outside formal healthcare settings and properly address this.

Exaggerated benefits of health AI

Without a doubt, AI applications in clinical practice and health research show significant promise. Yet, reading the AIDA report, one may get the impression that AI is widely and successfully used in clinical settings. Although the fight against COVID-19 has indeed accelerated research of new technologies in health, this has (so far) produced few robust, generalisable results. While some AI algorithms may perform with high accuracy on paper, even comparable to human specialists, they might not perform as well in real-world clinical practice. The expectations for AI in health appear inflated as the promised transformative events have for now mostly been confined to lab-controlled environments. Furthermore, the AIDA report seems to falsely equate more diagnoses with improved clinical outcomes. However, that is not always the case – overdiagnosis might even result in more harm than good. Ultimately, implementing AI technologies in health contexts requires accurate algorithms and investments in care infrastructure, professionals, and resources.

Underdiagnosis of the risks for both individuals and society

While the benefits of AI appear over-diagnosed, the AIDA report seems to downplay the risks. It correctly acknowledges some of them, in particular, harms to individuals’ wellbeing due to, for instance, misdiagnosis and related liability issues. But, by focusing on individual risks, the report overlooks broader societal risks and harms of AI. Unlike ‘human’ medical errors, mistakes of AI systems can have a much larger impact: technical system errors in AI could lead to mass patient injuries because of widespread use. Additionally, AI might perform better on sub-populations that are better studied or better represented in the training datasets, reflecting existing societal biases. For example, when AI is used to aid health professionals in diagnosing skin cancers, the system frequently uses freely available image databases that often do not include images of people of colour. As a result, these systems may lead to people of colour’s underdiagnosis. This would not only directly harm the health of individual members of (already) marginalised groups; it could also deepen the (existing) socio-economic inequalities in terms of access to health and health outcomes. It thus seems that the report disregards these societal implications in favour of potential economic benefits.

Patients are not mere data points; they are humans with fundamental rights

The risks of AI in health go well beyond health risks. AI runs on data, but data is not just an input; it is “collected by people, from people and for people”. Data subjects and those affected by AI are human beings with fundamental rights like dignity, non-discrimination and privacy. For instance, the report does not touch upon the black-box nature of many AI systems; a lack of understanding of how AI reaches its conclusion makes it difficult for clinicians to explain and for patients to understand AI’s advice, which can impair existing health practices of information provision and informed consent as an integral part of the right to health. Moreover, those seeking care are in a specific situation of vulnerability due to asymmetry in information and risk aversion. Therefore, the EU’s approach to AI in health should not only ensure health-specific fundamental rights are protected in the design, deployment, and usage of AI in health tools but also to address citizens’ willingness to provide and curate health data and ways to govern it appropriately.

Development of AI in health requires a specific approach

Developing health solutions is different from conventional product development because it touches upon people’s bodies and lives in a way no other industry does. Introducing AI in health amplifies the need for multidisciplinary teams that bring together insights from (bio)medicine, data science, behavioural sciences, and ethics and law. By lacking identification of this heterogeneity, namely through the recognition of the need for sector-specific regulation, the report fails to address the needs of patients, practitioners, and also developers and industry.

The AIDA report should therefore acknowledge the special nature of AI in the context of health and call for the proper regulation of the risks to patients or consumers as affected individuals and society. After all, health is everyone’s concern, and the algorithms have already entered doctor’s offices and our living rooms. Let’s ensure that they are really here to make us healthier.

This blog was first published on Euractiv on 24 January 2022.