Reclaiming Power over AI: Equipping Queer Teens as AI Designers for HIV Prevention
Project Overview
The document explores the role of generative AI (GenAI) in education, particularly its application in supporting HIV prevention initiatives targeted at LGBTQ+ adolescents. It emphasizes the transformative potential of GenAI tools in bridging information gaps and improving healthcare access for marginalized groups. The findings highlight both the opportunities and challenges associated with GenAI, particularly the risk of biased outputs that could worsen existing inequities. The authors advocate for the creation of queer adolescent-centered interventions and stress the importance of participatory frameworks that involve LGBTQ+ youth in the design and implementation of AI tools. Community engagement is underscored as a critical component in developing inclusive and effective solutions that address the unique needs of this demographic. Overall, the document calls for a thoughtful approach to the integration of GenAI in educational contexts, aiming to leverage its capabilities while mitigating risks to ensure equitable outcomes.
Key Applications
Generative AI tools like ChatGPT for personalized healthcare and HIV prevention education.
Context: Targeted towards LGBTQ+ adolescents to provide sexual health information and support.
Implementation: A community-engaged approach where queer teens with expertise in sexual health co-design GenAI health tools.
Outcomes: Improved access to sexual health information and greater engagement in HIV prevention strategies.
Challenges: Bias in AI models leading to misgendering and potential emotional distress for users.
Implementation Barriers
Technical Barrier
Advanced technical knowledge required for marginalized individuals to engage in AI design processes.
Proposed Solutions: Adopting frameworks like meta-design for resistance to empower users and facilitate their involvement.
Bias in AI
AI models trained on biased datasets can lead to harmful outputs for LGBTQ+ users.
Proposed Solutions: Implementing personalized guardrails and involving marginalized communities in AI tool design.
Project Team
William Liem
Researcher
Andrew Berry
Researcher
Kathryn Macapagal
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: William Liem, Andrew Berry, Kathryn Macapagal
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai