AI used by the state: Safeguarding autonomy and human rights with transparency to citizens and support for public servants

Artificial intelligence systems are becoming a key technology for the successful digitalization of public administration. At the same time, the introduction of AI systems in public administration raises relevant ethical and legal questions with regard to the legal protection and autonomy of individuals, as well as the integrity of democratic societies and the rule of law as a whole. Ensuring the greatest possible transparency regarding the use of AI systems is therefore of paramount importance. The Netherlands is considered a pioneer in this respect, having already introduced an AI transparency register. An idea that also has great potential for Germany. Together with civil society partners (AlgorithmWatch and NExT e. V. - Netzwerk: Experten für die digitale Transformation der Verwaltung), a viable framework for an AI transparency register for public administration and legal requirements for its implementation will therefore be developed during the one-year funding period. In particular, the requirements of the new European AI Act and the remaining regulatory leeway of the member states are to be taken into account. Based on an impact assessment, the register will provide information on the key characteristics of AI systems. This will not only benefit civil society, but also the administration, as it provides crucial information that was previously lacking.