A recent study launched by UNESCO has shed light on the prevalence of gender bias in leading artificial intelligence tools developed by OpenAI and Meta. The findings, released Thursday, raise concerns about the portrayal of women in AI-generated content and highlight the need for greater diversity and ethical oversight in the tech industry.
The study focused on examining algorithms such as Meta’s Llama 2 and OpenAI’s GPT-2 and GPT-3.5, commonly referred to as Large Language Models (LLM), that power popular AI applications including chatbots and content generators. According to UNESCO experts, the analysis revealed “unequivocal evidence of prejudice against women” in the outputs generated by these AI models.
Women were consistently associated with words related to domestic roles such as “home”, “family” or “children”, while men were associated with concepts such as “business”, “salary” or “career”. In addition, the study found that men were often depicted in prestigious professions such as teachers, lawyers, and doctors, while women were disproportionately represented as prostitutes, cooks, or domestic servants.
Although GPT-3.5 showed relatively less bias than the other two models, concerns were raised about its closed source, limiting transparency and accountability. Conversely, Llama 2 and GPT-2, which are open source, have allowed more control over the biases embedded in their algorithms. Leona Verdadero, UNESCO digital policy specialist, expressed concern about the impact of biased AI on users and stressed that technology companies must address these issues to effectively serve all users.
UNESCO Director-General Audrey Azoulay highlighted the significant influence of AI tools in shaping public perception and called for joint efforts to address gender bias in AI content. She emphasized the importance of promoting diversity in AI companies and called on governments to regulate the development and deployment of AI technologies to ensure ethical practices.
The release of UNESCO’s report coincides with International Women’s Day and serves as a timely reminder of the need to address gender bias in technology and promote equality and inclusivity in the digital age.