Document Type

Article

Publication Date

2023

Abstract

A mapping of recent Concluding Observations issued to States parties by the Committee on the Elimination of All Forms of Discrimination against Women (CEDAW), the Committee on the Rights of the Child (CRC), the Committee on the Rights of Persons with Disabilities (CRPD), and the Committee on the Elimination of Racial Discrimination (CERD) provides an analysis of the evolving human right standard by which to examine gender and intersectional stereotypes. The mapping exercise focused on cultural practices and gender stereotypes in the Concluding Observations reveals the CEDAW Committee is more likely than any other above-mentioned treaty body to discuss stereotypes. This provides us with a textual understanding of how stereotypes, culture and traditional practices often overlap and intersect. While gender stereotypes have replaced more overt forms of gender discrimination, these subtle stereotypes constitute different challenges as they are less visible to the untrained eye. The second part of the paper examines how a new generation of stereotypes are being baked into Artificial Intelligence (AI) through AI training data. While writing this paper, Open AI released ChatGPT and other Generative AI and Large Language Models which demand critical examination for emerging forms of gender bias. Last summer, Satya Nadella, the CEO of Microsoft, known for his love of poetry, previewed Open AI’s newest model on Generative AI and asked the chatbot to translate the Persian poet Rumi into Urdu, and then English. He recalls that he exclaimed: “God, this thing.” Despite the strides in AI innovation, the dangers of potential gender bias are real. Prometheus stole fire from the Greek Gods and was punished for his folly and hubris. He still provided humans with the fire of life and was pardoned by the great Zeus himself. As much as these new technologies have the potential for great good and can advance medicine, science, health care, food security and other forms of human endeavor, they also have great potential for harm and pose risks to human rights. The CEDAW’s new General Recommendation (GR) 40 and the GR 41 (in the pipeline) can create important new normative frameworks that propose a human right-based approach to mitigate the emerging gender stereotypes of new technologies.

Keywords

gender stereotypes, human rights, discrimination, cultural practices, new technologies, Artificial Intelligence, AI innovation, ChatGPT, gender bias, CEDAW, CERD, CRC, CRPD normative frameworks

Publication Title

Journal of Law & Public Affairs

DOI

https://doi.org/10.58112/jlpa.8-2.2

Download

Link to SSRN

Share

COinS