Gender-Sensitive AI Policy in India

Research Objectives

• Examine the current AI policy landscape in India, particularly initiatives led by NITI Aayog and MeitY.

• Investigate gender-related challenges experienced by women working in AI and STEM.

• Explore perceived algorithmic bias in applications such as recruitment, healthcare, financial services, and facial recognition.

• Evaluate how effectively existing AI governance frameworks address gender equality.

• Develop actionable recommendations for gender-responsive AI policies in India.

Research Methodology

The study adopted a mixed-methods research approach, combining quantitative research, qualitative interviews, and secondary policy analysis.

Quantitative Research

A structured questionnaire was designed for 50 women working or studying in AI and STEM-related fields. The survey examined professional barriers, perceptions of algorithmic discrimination, women's representation in AI governance, and awareness of national AI policies.

Qualitative Research

Five semi-structured expert interviews were incorporated to explore the relationship between AI development, gender inclusion, institutional accountability, and policymaking. The interviews were designed to provide contextual understanding of the issues investigated through the survey.

Secondary Research & Policy Analysis

A comparative review examined Indian AI strategies alongside international frameworks, including:

• NITI Aayog's National Strategy for Artificial Intelligence: #AIforAll (2018)

• Responsible AI for All (2021)

• UNESCO's Recommendation on the Ethics of Artificial Intelligence (2021)

• Gender-Sensitive AI Policy in Southeast Asia (2023)

• India's Digital Personal Data Protection Act (2023)

• CEDAW and international gender equality frameworks

Analysis Methods: IBM SPSS Statistics for quantitative analysis, thematic analysis for expert interviews, and comparative policy analysis.

Key Research Insights

The research identified four interconnected areas requiring attention:

1. Gender representation in AI development

   Women's participation in AI extends beyond workforce diversity. Representation in technical teams, leadership, and policymaking can influence whose needs are considered during technology development.

2. Algorithmic bias as a systemic challenge

   Gender bias can enter AI systems through unrepresentative training datasets, historical discrimination, design decisions, and inadequate testing. These risks are particularly consequential in high-impact applications such as hiring, healthcare, and financial services.

3. The gap between ethical principles and implementation

   India's AI strategies emphasize fairness, inclusion, and responsible innovation. However, translating these principles into measurable gender-specific safeguards remains an important governance challenge.

4. The need for context-sensitive AI governance

   India's social, economic, linguistic, and regional diversity requires AI governance approaches that account for intersecting inequalities rather than treating women as a homogeneous population.

Research Recommendations

The study proposes a gender-responsive approach to AI governance built around five interventions:

• Gender impact assessments: Introduce structured assessments and bias audits for high-risk AI applications.

• Participatory governance: Increase women's representation in AI policymaking, regulatory discussions, and ethics committees.

• Inclusive datasets: Establish standards for representative data collection, evaluation, and algorithmic fairness.

• Capacity building: Expand AI education, mentorship, research opportunities, and entrepreneurship support for women.

• Institutional accountability: Develop mechanisms for monitoring discriminatory outcomes and evaluating the effectiveness of AI governance measures.

Research Contribution

This project connects human-centered design, technology ethics, gender studies, and public policy to examine AI as a sociotechnical system rather than a purely computational technology.

By combining stakeholder perspectives with policy analysis, the research offers a framework for understanding how gender inclusion can be embedded throughout the AI lifecycle—from data collection and model development to deployment and governance.

The central argument is that achieving India's vision of “AI for All” requires moving beyond broad commitments to inclusion toward measurable, accountable, and gender-responsive AI governance.

Research Collaboration: Dr. Harshini Raji VP and Dr. Uma Maheswari P, Department of Media Sciences, Anna University, Chennai.