"As artificial intelligence (AI) systems increasingly influence critical sectors such as healthcare, employment, education and law enforcement, concerns around bias – especially gender bias – have come to the forefront. Gender bias in AI not only reflects but can escalate existing inequalities, raising significant ethical, legal and societal is sues. This policy paper examines the impact of gender bias in AI systems and presents comprehensive guide lines for addressing it through a socio-technical lens. By focusing on different stages of the AI lifecycle, the paper provides actionable recommendations for various stake holders, including developers, deployers, users and regulators. Therefore, the aims of this document are to: → raise awareness about the escalation of gender bias when using AI systems in the decision-making process; → outline the challenges and opportunities of incorporating a socio-technical approach to tackle gender bias issues in AI systems; → provide a set of recommendations to key stakeholders from a socio-technical perspective on how to identify and prevent the reproduction of gender bias..."

Support: Digital