Abstract:
The verification of digital social systems at the governmental level of individual countries leads to the reflection that they enjoy a clear interest of the state administration. At the same time it cannot be overlooked that the increasing number of digital instruments in the broadly understood social
sphere is also provoking an avalanche of negative reactions. A number of scientists and non-governmental organisations in the context of the technologies used point to the threat to privacy and individual autonomy, and moreover to the arbitrariness of decisions, discriminatory behaviour of
administrative bodies and excessive interference in civil rights and freedoms.
Therefore, continual analysis of digital solutions in the area of social assistance and social care is not so much the need, but the necessity. Such an analysis should include definition of objectives, methods and nature of algorithmic tools used. It should also include a review of normative solutions regulating the process of design, implementation, use and control of algorithmic systems, evaluation of their use and definition of threats. Compromise solutions in this area include the establishment of control bodies and the use of machine
learning models, whose activities will be subject not only to ex post supervision, but also to ex ante verification in the design phase.
