Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

"OpenBloom": A Question-Based LLM Tool to Support Stigma Reduction in Reproductive Well-Being

Ashley Hua, Adya Daruka, Yang Hong, Sharifa Sultana

TL;DR

OpenBloom investigates how stigma around reproductive wellbeing interacts with AI-generated educational prompts. Treating OpenBloom as a design probe, the study combines surveys, interviews, and focus groups to examine interactional dynamics, not stigma reduction alone, revealing that current LLM outputs tend toward superficial rephrasing rather than deep, values-based reflection. Key contributions include design recommendations for empathetic framing, inclusive representation, and explicit engagement with marginalized identities, plus a feminist HCI lens (Bardzell) to critique the implementation gap and propose participatory, human-in-the-loop workflows. The work highlights the necessity of culturally grounded, socially accountable AI design in sensitive health domains and argues for ongoing community involvement to avoid reinforcing inequities. The practical impact is guidance for building stigma-sensitive reproductive health education tools that balance accuracy with cultural relevance and ethical considerations."

Abstract

Reproductive well-being education remains widely stigmatized across diverse cultural contexts, constraining how individuals access and interpret reproductive health knowledge. We designed and evaluated OpenBloom, a stigma-sensitive, AI-mediated system that uses LLMs to transform reproductive health articles into reflective, question-based learning prompts. We employed OpenBloom as a design probe, aiming to explore the emerging challenges of reproductive well-being stigma through LLMs. Through surveys, semi-structured interviews, and focus group discussions, we examine how sociocultural stigma shapes participants' engagements with AI-generated questions and the opportunities of inquiry-based reproductive health education. Our findings identify key design considerations for stigma-sensitive LLM, including empathetic framing, inclusive language, values-based reflection, and explicit representation of marginalized identities. However, while current LLM outputs largely meet expectations for cultural sensitivity and non-offensiveness, they default to superficial rephrasing and factual recall rather than critical reflection. This guides well-being HCI design in sensitive health domains toward culturally grounded, participatory workflows.

"OpenBloom": A Question-Based LLM Tool to Support Stigma Reduction in Reproductive Well-Being

TL;DR

OpenBloom investigates how stigma around reproductive wellbeing interacts with AI-generated educational prompts. Treating OpenBloom as a design probe, the study combines surveys, interviews, and focus groups to examine interactional dynamics, not stigma reduction alone, revealing that current LLM outputs tend toward superficial rephrasing rather than deep, values-based reflection. Key contributions include design recommendations for empathetic framing, inclusive representation, and explicit engagement with marginalized identities, plus a feminist HCI lens (Bardzell) to critique the implementation gap and propose participatory, human-in-the-loop workflows. The work highlights the necessity of culturally grounded, socially accountable AI design in sensitive health domains and argues for ongoing community involvement to avoid reinforcing inequities. The practical impact is guidance for building stigma-sensitive reproductive health education tools that balance accuracy with cultural relevance and ethical considerations."

Abstract

Reproductive well-being education remains widely stigmatized across diverse cultural contexts, constraining how individuals access and interpret reproductive health knowledge. We designed and evaluated OpenBloom, a stigma-sensitive, AI-mediated system that uses LLMs to transform reproductive health articles into reflective, question-based learning prompts. We employed OpenBloom as a design probe, aiming to explore the emerging challenges of reproductive well-being stigma through LLMs. Through surveys, semi-structured interviews, and focus group discussions, we examine how sociocultural stigma shapes participants' engagements with AI-generated questions and the opportunities of inquiry-based reproductive health education. Our findings identify key design considerations for stigma-sensitive LLM, including empathetic framing, inclusive language, values-based reflection, and explicit representation of marginalized identities. However, while current LLM outputs largely meet expectations for cultural sensitivity and non-offensiveness, they default to superficial rephrasing and factual recall rather than critical reflection. This guides well-being HCI design in sensitive health domains toward culturally grounded, participatory workflows.
Paper Structure (36 sections, 3 figures, 1 table)

This paper contains 36 sections, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. AI in Sexual and Reproductive Health
  4. Cultural Sensitivity in Health AI
  5. AI-Generated Questions for Learning
  6. Reproductive Health Stigma and HCI
  7. Methods
  8. Application Design and Functionality
  9. Study Procedures
  10. Survey
  11. Individual Interviews
  12. Focus Group Discussions
  13. Data Analysis
  14. Findings
  15. Reproductive Health Stigma: Manifestations, Myths, and Marginalization
  16. ...and 21 more sections

Figures (3)

  • Figure 1: Work-flow diagram of OpenBloom, from article submission to AI-generated summarization, question generation, and response grading.
  • Figure 2: (a) Landing Page for OpenBloom, (b) Article summary generated by OpenBloom, (c) Excerpt of questions generated by OpenBloom, and (d) Scores shown upon taking the quiz.
  • Figure 3: Survey results: (a) Frequency of responses for each survey question, (b) Mean Likert scores (1–5).