Like many other areas, the HR industry is undergoing a process of automation and algorithmisation. AI-supported technologies are increasingly being researched and used for tasks such as contacting applicants, filtering CVs or conducting job interviews. At the same time, a shortage of skilled workers in the state of Baden-Württemberg is rendering it increasingly difficult to fill positions in areas such as mechanical engineering or mechatronics. The FAIR project, funded by the Invest BW program of the Ministry of Economics, Labour and Tourism, aims to develop a fair, diversity-sensitive and transparent AI system in order to attract highly qualified candidates specifically for companies in Baden-Württemberg.

A central claim made by providers of AI technology is that – unlike human actors – it is unbiased. But given that AI systems rely on human inputs and data from complex social contexts, they can also reproduce and automate social biases. When used in HR, they can disadvantage vulnerable groups such as ethnic minorities, women or people with disabilities. In addition, the underlying algorithmic decisions are often opaque and hardly interpretable to humans.

The work of the IZEW therefore plays a key role in FAIR. Developing a system that is as non-discriminatory and transparent as possible requires a continuous examination of ethical norms beyond technical requirements for algorithmic fairness, explainability and diversity. The IZEW accompanies the project development and ensures the inclusion of ethical aspects in the context of recruiting and personnel selection. Together with Reutlingen University, the Ethics Centre analyses the individual and social acceptance of AI software and develops criteria for an ethically acceptable use. Ethically informed specifications and training materials for users are developed in close collaboration with the project partners. The research has the potential to set ethical standards for future studies and products in AI-assisted recruiting.