January 1, 2024

Systemic Prejudices - An Investigation into Bias Against Women and Girls in Large Language Models

UN Entity:

UNESCO

SDGs:

SDG 05: Gender Equality

Innovation Area:

Artificial Intelligence & Machine Learning
This study explores biases in three significant large language models (LLMs): OpenAI’s GPT-2 and ChatGPT, along with Meta’s Llama 2, highlighting their role in both advanced decision-making systems and as user-facing conversational agents. Across multiple studies, the brief reveals how biases emerge in the text generated by LLMs, through gendered word associations, positive or negative regard for gendered subjects, or diversity in text generated by gender and culture. 
View RecordingView Resource
UNESCO

More From the Library

August 15, 2022
August 15, 2022

Go.Data Gitub Documentation Site

Go.Data is an outbreak investigation tool for public health emergencies, and includes features for contact tracing, contact follow-up, and visualizing chains of transmission. This repository helps users implement Go.Data for outbreak investigation.
September 22, 2023
September 22, 2023

Why Generative Methods Are Still Missing Something to Become Responsible

This talk comments on Responsible AI in a multi-dimensional approach and delves into the complexities surrounding the apparent oversight of the human factor in these impressive adaptive algorithms.
June 22, 2023
June 22, 2023

A Principled Framework for Responsible Data Sharing Between Humanitarian Organizations and Donors

The framework offers six guidelines to inform data sharing between humanitarian organizations and donors.
August 15, 2022
August 15, 2022

Digital Inclusion of Refugees Resettling to Canada: Opportunities & Barriers

This report provides insight into how refugees who participate in the Canadian Orientation Abroad programme are using digital tools in the pre-departure stage, and what challenges they face in doing so.
Back to the Innovation Library
WIPO
UN Tourism
WTO
UPU
UNESCO
Publication
UNESCO
Artificial Intelligence & Machine Learning

Join the UN Innovation Network

Thanks for your interest in joining the UN Innovation Network (UNIN). UNIN is a network for UN employees interest in innovation; we welcome non-UN colleagues in an observer role.

To get to know our members, their skills and expertise better, we would like to ask you a few questions about your work and interests. Please complete this form to reflect your own experiences and interests, not those of your team or organisation.

More From the Library

No items found.