Council of Europe's Work in progress
Updated on 2 July 2019
The European Commission against Racism and Intolerance and Inclusion and Anti-Discrimination Programmes published the study “Discrimination, Artificial Intelligence and Algorithmic Decision-Making”, prepared by Prof. Frederik Zuiderveen Borgesius for the Anti-discrimination Department of the Council of Europe. The study elaborates on the risks of discrimination caused by algorithmic decision-making and other types of artificial intelligence (AI). The study formed basis of statement made by Christian Ahlund, Bureau member of ECRI, on Artificial Intelligence and Democracy, at the High Level Conference on AI, Helsinki, 26 February 2019. Discussions were held at EuroDIG 2019, pre event involving the Finish Ombudsperson, Amnesty International and Internet start-ups. ECRI will host a session on AI and Discrimination during its 25th anniversary conference in September in Paris. Via its field programmes on anti-discrimination and hate speech, awareness-raising modules and tools for equality bodies will be developed.
The Committee on Bioethics (DH-BIO) is preparing a Strategic Action Plan on technologies and human rights in the field of biomedicine 2020-2025, which includes focus on the protection of vulnerable groups such as children and the elderly. A Guide on public engagement on fundamental questions raised by developments in biology and medicine with particular focus on new technologies is also under preparation.
The Council of Europe Strategy for the rights of the child (2016-2021) includes a focus on children’s rights on the Internet, and actions implemented take into account that new technologies, such as artificial intelligence (AI), inevitably impact on children’s wellbeing and children’s rights. This dimension has been fully considered by the Ad hoc Committee for the Rights of the Child (CAHENF) in the preparation of the Recommendation CM/Rec(2018)7 of the Committee of Ministers to member States on Guidelines to respect, protect and fulfil the rights of the child in the digital environment. An Implementation handbook is currently under preparation and will include further guidance and examples of good practices, including on AI related issues.
The Internet Literacy Handbook is a tool for children, parents, teachers and policy makers to be able to make the most of the Internet and prepare children to use the Internet safely and confidently by being aware of opportunities and risks. The translation and availability of the Internet Literacy Handbook in additional languages and AI related factsheets for use by teachers, educators, parents and students (2018-2019) is ongoing.
The Culture and Cultural Heritage Division organised a seminar on culture, creativity and artificial intelligence for experts and CDCPP members within the framework of the Croatian Chairmanship of the Committee of Ministers of the Council of Europe (E-relevance of Culture in the Age of AI, Rijeka, 12-13 October 2018). The event generated a set of examples and guidelines on how policies could be shaped to address the challenges and opportunities of AI in this sector, for member States and other stakeholders. A comprehensive publication bringing together essays and reflections by leading AI specialists in the cultural and creative field is forthcoming. A short booklet was already published. Video testimonials and conclusions are also published online. All the news are published on the dedicated webpage.
The Consultative Committee of Convention 108 published guidelines on the data protection implications of artificial intelligence which provide a set of baseline measures that governments, AI developers, manufacturers, and service providers should follow to ensure that AI applications do not undermine the human dignity and the human rights and fundamental freedoms of every individual, in particular with regard to the right to data protection with recommendations (regarding possibilities to limit their impact on dignity and individual freedom and the importance of taking into account the ethical dimension of the use of such technologies).
The modernised Convention 108 provides new protective means for the individuals facing algorithmic decision making
Guidelines on big data were adopted by the Committee in 2017 which lay the foundations for the protection of individuals with regard to the processing of personal data in a big data era.
The Democratic Governance Department is currently working on E-Voting and European Electoral Performance Index.
As regards to E-Voting and on the basis of Recommendation (2017)5 on E-Voting, there will be a review of the areas where AI may produce impact throughout the entire course of the E-Voting process and proposals of solutions to maximize positive and minimize negative effect of AI in this respect.
In the framework of the European Electoral Performance Index (EEPI), there will be a research of the criteria and the influence of various forms of artificial intelligence and automated algorithms which might have relevance (in addition to the existing indicators) and which may have an impact (positive or negative) on the electoral cycles.
The Education Department will hold an Expert Seminar within the framework of the project Digital Citizenship Education with a view to examine the challenges of AI for policy makers in Education – from e-learning to intelligent learning. Producing a policy paper – in view of the preparation of a Digital Citizenship Education Strategic Plan.
Promoting freedom of expression and a diversified cultural offer represent core goals for Eurimages. This is the reason why the Fund is closely following the shifts in consumption habits in the European film industry. New technologies and AI have changed the rules of the game and impacted cultural diversity. For better or for worse.
During the past few years, video on Demand (VoD) digital platforms, mainly non-European, have represented the fastest audience growth in comparison with cinemas and television. As for example, Netflix is quickly approaching 150 million subscribers in over 190 countries.
Thanks to predictive technologies based on artificial intelligence and big data, these platforms have developed very powerful recommendation systems. The majority of streaming subscribers choose what to watch as a result of a decision made by one or several algorithms. More than 80 per cent of the content watched on Netflix is discovered through the platform’s recommendation system.
In addition to content distribution, the streaming giants have also become important protagonists in the funding of audiovisual works. Their investments in the production of content are also driven by “data analytics” and the preferences of the platforms’ users. The model developed is centred on demand (bottom-up) and no longer on supply (top-down).
These structural changes in both production and distribution raise many questions at ethical, cultural, economic and political levels.
The aim of the study launched in May 2019 is to establish whether the current situation of the audiovisual sector in our member states requires putting in place specific measures to ensure that the Council of Europe’s core values are upheld.
The Group of States against Corruption (GRECO), which is the Council of Europe anti-corruption body, has yet to analyse in depth all the implications of AI on corruption but has started ground work to explore risks and benefits of AI. Firstly, measures need be taken to prohibit the use of software algorithms with corrupt intent. Knowing how easily computer programmers can alter search algorithm to direct people to one content over others or gather information from individual profiles on social media, persons with wrong intentions obviously take great interest in how to use algorithms to their advantage. At the same time, the great potential of AI for the fight against corruption should not be downplayed. AI can improve the effectiveness of steps taken to combat corruption, for instance in being used in investigations into complex, lengthy corruption cases or in helping compliance officers identify corruption risks or offenses. If the fight against corruption is to truly benefit from the full potential of AI, some safeguards need therefore to be put in place to avoid any misuse.
Fight against Crime (other than cybercrime)
The European Committee on Crime Problems (CDPC) is preparing a report on substantive criminal law challenges posed by advances in robotics, artificial intelligence and smart autonomous machinery, including self-driving cars, drones and other forms of robots capable of causing physical harm independent of human operators. Possible standard-setting instrument addressing that concern might take the form of a Council of Europe convention.
A plenary session about the implications for criminal justice of AI has been organised during the Octopus Conference 2018. Some videos with experts’ comments were published.
The Gender Equality Commission (GEC) adopted a draft Committee of Ministers Recommendation to prevent and combat sexism in December 2018. This includes guidelines on how to avoid the potential risks of technology integrating and strengthening existing gender biases and looking at how AI could help to close gender gaps and eliminate sexism (see part II.B.7 of the Recommendation).
At its meeting on 22-24 May, the GEC also identified AI and gender equality as a possible topic for a thematic section of its 2019 Annual Report on implementation of the Gender Equality Strategy, and as a priority area for further study during the 2020-2021 biennium. The GEC is particularly interested in contributing towards the development of a Council of Europe legal framework on AI.
Under the supervision of the Steering Committee on Media and Information Society (CDMSI), the committee of experts on internet intermediaries (MSI-NET) has realised in 2017 a study on the human rights dimensions of automated data processing techniques (in particular algorithms) and possible regulatory implications.
As a follow-up, the Committee of experts on human rights dimensions of automated data processing and different forms of artificial intelligence (MSI-AUT) is now preparing draft recommendation on the human rights impacts of algorithmic systems. The Committee is also preparing a study on the implications of AI on the concept of responsibility within a human rights framework.
In February 2019, the information society department organised in cooperation with the Finnish Government and under the Finnish Chairmanship of the Committee of Ministers, a high-level conference 'AI: Governing the Game Changer', bringing together experts and policy makers from multiple disciplines to discuss the impacts of AI development on human rights, democracy and the rule of law, and to explore options for coordinated action towards ensuring that adequate controls and democratic oversight are in place.
The European Commission for the Efficiency of Justice (CEPEJ) adopted, on 3 December 2018, the first European Ethical Charter on the use of artificial intelligence in judicial systems, setting out 5 principles which should guide the development of AI tools in the European judiciaries. The Charter included several annexes: a scientific study, a glossary and a review of different AI applications with recommendations on their use. Through this work, CEPEJ intended to promote a responsible use of AI, driven by respect for fundamental rights and ethical principles and meeting the needs of legal professionals.
In 2019, efforts will be deployed to make sure that the Charter becomes "a living instrument" within European judiciaries, in line with the orientations provided in the document CEPEJ-GT-QUAL (2019)1. The CEPEJ will in particular carry out a work of further development of the principles of the Charter and examine the feasibility of a certification of AI tools and services, against the Charter's principles. The CEPEJ Secretariat stands ready to accompany private and public actors in the application of the Charter's principles.
The CEPEJ co-organised with the Court Administration of the Republic of Lavia a conference on “Artificial intelligence at the service of the judiciary” in Riga, on 27 September 2018. The conference gathered Latvian judicial authorities and representatives from a number of European countries and allowed reviewing different initiatives being taken to integrate artificial intelligence in judicial policies, as well as discussing the main challenges and opportunities arising in this respect.
The CEPEJ Secretariat will organize, upon request of member States, specific activities to facilitate the implementation of the European Ethical Charter. In this vein for instance, the CEPEJ Secretariat hold a mission to the UK on 14 February; in this connection, the CEPEJ will be audited by the Law Society of England and Wales which is drafting a report on algorithms in the UK justice system.
The European Committee on Legal Co-operation (CDCJ) is currently working on ODR (online dispute resolution) mechanisms and is taking into consideration possible applications of AI in such systems and their compliance with the right to a fair trial and the right to an effective remedy. A drafting goup of its members will be set up in 2019 to develop draft guidelines for the attention of policy-makers responsible for designing online dispute resolution mechanisms with a view to ensuring the compatibility of such mechanisms with Articles 6 and 13 of the ECHR.