Back to Labour and employment

Article 14 and Protocol No. 12 ECHR prohibit discrimination in employment-related matters on a wide range of grounds. Gender-based discrimination in employment-related matters in particular is a recurrent problem: “where a difference of treatment is based on sex, the margin of appreciation afforded to the State is narrow and in such situations the principle of proportionality does not merely require that the measure chosen should in general be suited to the fulfilment of the aim pursued, but it must also be shown that it was necessary in the circumstances”.[1]  The advancement of gender equality is today a major goal in the member States of the Council of Europe and very weighty reasons would have to be put forward before such a difference of treatment could be regarded as compatible with the ECHR.[2]

As regards the ESC, to comply fully with the rights to access employment (Article 1§2),[3] to equal pay between women and men (Article 4§3),[4] and to equal opportunities in matters of employment (Article 20)[5], States Parties must implement legal measures to ensure the effective enforcement of the prohibition of discrimination. Should claims of discrimination arise, effective remedies include judicial and administrative procedures, ensuring access to reinstatement, compensation, and enforceable penalties, with labour inspections playing a key role in enforcement.[6] These remedies must be adequate, proportionate, and dissuasive to ensure meaningful protection against discrimination.[7]

AI systems are increasingly being used in recruitment processes,[8] which may be negatively affected, for example in cases where reliance on machine learning in the identification of candidates led to discriminatory outcomes, or where AI-based facial recognition and emotion analysis systems have resulted in racial discrimination.[9] As such, AI systems used for recruitment and selection of candidates should be objective, neutral and free from bias, including gender bias. In a broader context, States should ensure that the use of AI systems in the workplace does not reproduce or amplify existing patterns of inequality and promotes equality including gender equality, diversity and inclusion. In particular, this could consist of regular auditing of the outcomes of the use of AI systems in recruitment, promotion and other procedures; the involvement of employees and their representative organisations in policies or choices regarding the use of AI in decision-making in the workplace; monitoring of the impact of the introduction of AI systems in the workplace on gender equality and diversity in the workforce; and training and awareness-raising for the workforce on data bias, stereotypes and risks of discrimination in using AI systems.

The use of AI in employment also harbours the risk that inequalities will persist and worsen if, for example, older persons, persons with disabilities or persons with limited digital skills lack the required abilities to use them. For these people, the chances of participating in the labour market may deteriorate. In addition, limited access to AI systems and tools can prevent individuals from experiencing the benefits and advantages which they may offer. Policymakers should ensure that AI systems are accessible, and promote together with employers the development and diffusion of AI systems that contribute to better participation in employment, such as assistance systems.

Back to Labour and employment