Can We Build AI That Does Not Harm Queer People?

This article examines the harmful information generated by artificial intelligence (AI) regarding aspects of the queer community with a focus on what can be done by computer scientists to prevent this problem. First, specific examples are detailed with a look at how they occur then a case study is p...

Descripción completa

Detalles Bibliográficos
Publicado en:	Communications of the ACM Vol. 68; no. 5; pp. 21 - 24
Autor principal:	Weber, Sabine
Formato:	Opinion
Publicado:	Association for Computing Machinery May2025
Materias:	LGBTQ+ communities Artificial intelligence Machine learning Computer science education Employee training Computer programmers Language models
Acceso en línea:	Ver este registro en EBSCOhost

Descripción
Sumario:	This article examines the harmful information generated by artificial intelligence (AI) regarding aspects of the queer community with a focus on what can be done by computer scientists to prevent this problem. First, specific examples are detailed with a look at how they occur then a case study is presented. A fictitious software company’s text-to-image system and how software developers can work to overcome harm with direct and indirect methods is the focus on the case study.