Committee of Ministers: Warning on risk of algorithms being used to manipulate social and political behaviour

IRIS 2019-4:1/3

Léa Chochon

European Audiovisual Observatory

In a Declaration adopted on 13 February 2019, the Committee of Ministers of the Council of Europe warned its 47 member states against the risk presented by algorithmic processes, particularly micro-targeting techniques, with regard to individuals’ decision-making and the formation of opinions.

Public awareness remains very limited regarding the extent to which everyday devices generate and use vast amounts of data. Apart from the notion of the protection of such data, the way in which it is used to train machine-learning technologies presents serious risks for the process of forming opinions and taking decisions, particularly due to a lack of awareness among the general public and the stakeholders, and an insufficient number of more in-depth studies on the subject. Data is used by these technologies to prioritise search results, from which it is possible to deduce individuals’ personal preferences, the information flows to which they have access, and the way in which they use their data; sometimes it is also used to subject individuals to behavioural experimentation. Moreover, the fine-grained and personalised levels of algorithmic persuasion are used not only to categorise people, but also increasingly to influence their emotions and thoughts so that their very ability to form an opinion and take a decision independently are threatened.

In view of all this, the Committee of Ministers:

- draws attention to the growing threat to the right of human beings to form opinions and take decisions independently of automated systems;

- encourages member states to assume their responsibility to address this threat at senior level, for example by using additional protective frameworks related to personal data;

- acknowledges the need to consider, at both national and international levels, the growing onus on the industry,

- stresses the role of academia in producing research on the capacity of algorithms tools to enhance or interfere with the cognitive sovereignty of individuals;

- recommends an appropriate assessment of the regulatory frameworks for political communication and election procedures and the possible need to adopt additional measures in order to guarantee the appropriate, democratic supervision of the development and use of algorithmic tools.

These issues will be covered in the near future at a Council of Europe conference in Helsinki on 26 and 27 February 2020 on the theme “Governing the Game Changer - Impacts of artificial intelligence development on human rights, democracy and the rule of law”.