About algorithmic behaviour control and artificial intelligence
The terms algorithms and artificial intelligence are omnipresent buzzwords. The underlying technologies are being used more and more frequently. At the same time they entail both risks and opportunities. Algorithms contain those control systems, which can be decisive for which conclusion can be made of the data processing and data analysis, predictions and decisions about people. They are no longer used solely for the automation of data processing and the generation of information and knowledge, but are increasingly core elements of systems for automated decision support or execution. Therefore, it is particularly important that software developers and clients implement value concepts and basic assumptions in algorithms, which can have both intended and unintended consequences for those directly affected as well as for third parties.
Hence, GOAL investigates how algorithms perform governance functions and how the governance of algorithms can be configured. The term „governance“ is used from a social science perspective and refers to collectively binding, societal forms of decision-making and control that are realized and enforced by more actors than the state alone. The focus is on the behavioural and regulatory effects of algorithms and the related social consequences.
The aim of GOAL is to identify governmental, technical and regulatory needs and options for action to design comprehensive governance structures. To this end, computer scientists, legal scholars, ethicists, behavioural scientists and economists are conducting joint interdisciplinary research on technology assessment in the GOAL project network. By bringing together different disciplines, new approaches to existing research problems can be developed. Events such as a winter school and an international workshop will also enable external experts to contribute to the project and integrate existing international knowledge. In addition, a case study with a self-developed app offers the opportunity to test the result of GOAL for the development of comprehensive governance structures in an exemplary manner and to demonstrate their suitability.
In a nutshell, the governance options elaborated by these interdisciplinary and practice-oriented approaches can have a significant influence on future technical developments.