The AI Gender Gap Paradox (SSIR) – Stanford Social Innovation Review

Report on the Gender Gap in Generative AI and Implications for Sustainable Development Goals

Executive Summary

A significant gender gap exists in the adoption and use of generative Artificial Intelligence (AI), posing a substantial threat to progress on several Sustainable Development Goals (SDGs), particularly SDG 5 (Gender Equality), SDG 8 (Decent Work and Economic Growth), and SDG 10 (Reduced Inequalities). This report analyzes the root causes of this gap, framing women’s hesitancy not as risk aversion but as “risk awareness” rooted in rational concerns about AI systems. It outlines the economic and social stakes and proposes a strategic framework for fostering inclusive AI adoption that aligns with the 2030 Agenda for Sustainable Development.

The Generative AI Gender Gap: A Barrier to SDG 5

Data indicates a persistent disparity in generative AI usage between men and women. This gap undermines the core principles of SDG 5 (Gender Equality) by limiting women’s access to transformative technology and perpetuating digital exclusion.

Evidence of the Divide

• A Harvard Business School meta-analysis of 18 studies found that women had 22 percent lower odds of using generative AI than men.
• Research from Deloitte indicates the gender gap is most pronounced in the 45+ age group, a demographic where women are often in senior leadership positions.
• Studies from Pew and Deloitte show women are more likely than men to predict that AI will cause more harm than benefit, reflecting a deep-seated skepticism.

Rational Basis for “Risk Awareness”

Women’s cautious approach is not unfounded but is based on legitimate concerns that must be addressed to ensure technology is developed inclusively, a key tenet of SDG 9 (Industry, Innovation, and Infrastructure).

1. Knowledge Gaps and Lack of Transparency: Women frequently cite the opaque nature of AI as a barrier to trust, hindering progress toward inclusive lifelong learning opportunities as outlined in SDG 4 (Quality Education).
2. System Unreliability: The propensity of AI models to “hallucinate” or generate fabricated information erodes trust and raises doubts about their utility.
3. Expectations of Sanction: Research shows that women are judged more harshly for using AI in the workplace, creating a professional penalty that directly contravenes the goal of full and productive employment for all under SDG 8.
4. Biased Outputs: AI systems trained on biased data perpetuate harmful stereotypes and discriminatory outcomes. Examples include AI advising women to ask for lower salaries, which directly widens the gender pay gap and works against SDG 5 and SDG 10 (Reduced Inequalities).
5. Privacy and Security Concerns: Women experience higher rates of tech-facilitated abuse and are consequently more concerned about data privacy, an issue central to building safe and inclusive digital spaces.

Economic and Social Implications for Sustainable Development

Failure to address the AI gender gap carries significant risks for global development objectives, potentially reversing progress on gender equality and economic inclusion.

Threats to SDG 8: Decent Work and Economic Growth

• Widening Pay Gaps: If men disproportionately benefit from AI-driven productivity gains, the existing gender pay gap is likely to widen, undermining Target 8.5 of SDG 8.
• Career Stagnation: Disengagement from AI tools can erode the competitive advantage of senior professional women, limiting their promotion opportunities and reducing female representation in leadership.
• Job Displacement: Predictions suggest women are significantly more likely to hold jobs that can be automated by AI, creating a disproportionate risk of economic disruption and challenging the goal of decent work for all.

Challenges to SDG 10: Reduced Inequalities

The lack of gender diversity in AI development teams leads to products that can feel foreign or hostile to women. This exclusion from the design and governance of technology exacerbates inequalities, as biased systems can reinforce discriminatory patterns in employment, housing, and finance, directly opposing the mission of SDG 10.

A Strategic Framework for Inclusive AI Aligned with the SDGs

To mitigate these risks, a multi-level strategy is required, drawing lessons from global development programs to turn women’s risk awareness into a force for creating more equitable and effective technology.

Recommendations for Action

1. Promote Community-Based Learning (SDG 4 & SDG 5): Establish peer-to-peer support networks where women can learn and experiment with AI in trusted, collaborative environments, building confidence and digital literacy.
2. Ensure Flexible and Accessible Design (SDG 10): Create inclusive learning platforms and communities that accommodate women’s time constraints and provide low-cost, on-demand training to reduce barriers to entry.
3. Foster AI-Powered Entrepreneurship (SDG 5 & SDG 8): Pair AI skills training with entrepreneurship and leadership programs to create new pathways for flexible income generation and economic empowerment for women.
4. Elevate Role Models and Leadership (SDG 5): Showcase women who are successfully using and building AI solutions to inspire others and foster mentorship networks that encourage engagement.
5. Enhance Transparency and Trust (SDG 16): Encourage AI companies to adopt robust transparency measures regarding data use, bias mitigation, and safety testing to build trust with skeptical users, contributing to more accountable institutions.
6. Establish Regulatory Oversight (SDG 16): Implement strong regulations, such as an AI Civil Rights Act, to protect consumers from biased systems, mandate audits, and ensure corporate accountability, thereby strengthening institutions that serve the public good.

Conclusion: Reclaiming the Narrative for a Sustainable Future

The narrative that mischaracterizes women as “risk-averse” must be replaced with an understanding of their “risk awareness” as a valuable asset. This perspective is crucial for identifying and rectifying flaws in AI systems, making the technology stronger, safer, and more equitable. Achieving the Sustainable Development Goals in an era of rapid technological change requires a commitment to fierce ambivalence—simultaneously leveraging AI for empowerment while demanding the highest standards of fairness and safety. By taking concerted action across learning, access, opportunity, accountability, and oversight, the global community can ensure that the AI revolution expands opportunity rather than deepening inequality.

Analysis of SDGs, Targets, and Indicators

1. Which SDGs are addressed or connected to the issues highlighted in the article?

1. SDG 5: Gender Equality
   • The entire article is centered on the gender gap in the adoption and development of generative AI. It discusses how women are underrepresented in AI development teams, leading to biased and hostile products. It also highlights the risk of AI widening the gender pay gap and reducing promotion opportunities for women.
2. SDG 8: Decent Work and Economic Growth
   • The article explores the impact of generative AI on the workplace, including significant productivity gains (“saved between 2-4 hours” weekly). It addresses the threat of job automation, noting that “women are nearly three times more likely than men to be in jobs that generative AI can easily automate.” It also points to new economic opportunities, such as using AI as a “springboard for solo-preneurship.”
3. SDG 10: Reduced Inequalities
   • The article explicitly warns that the gender gap in AI adoption could “widen existing inequalities.” It points out that biased AI systems can lead to “exclusionary outcomes for marginalized groups seeking housing, credit, or seeking employment opportunities,” extending the issue beyond gender to broader societal inequalities.
4. SDG 9: Industry, Innovation, and Infrastructure
   • The focus is on a major technological innovation—generative AI. The article calls for improving this technology by making it safer, more transparent, and less biased. It argues that women’s “risk awareness” can “strengthen the technology itself,” contributing to more resilient and inclusive technological development.
5. SDG 4: Quality Education
   • A significant portion of the article discusses the need for education and upskilling. It highlights that women desire AI training to overcome the technology’s opacity and build trust. The proposed solutions heavily feature “community-based learning and peer support,” “on-demand AI trainings,” and skills-building programs to ensure women can effectively use these new tools.
6. SDG 16: Peace, Justice and Strong Institutions
   • The article calls for accountability and oversight in the AI industry. It advocates for “exacting standards of transparency, fairness, and safety” and mentions the need for regulatory frameworks like the “Artificial Intelligence (AI) Civil Rights Act” to protect consumers, systematize bias surveillance, and create rules to “prevent corporate abuse.”

2. What specific targets under those SDGs can be identified based on the article’s content?

1. Under SDG 5 (Gender Equality):
   • Target 5.5: Ensure women’s full and effective participation and equal opportunities for leadership. The article warns that if experienced women disengage from AI, “boardrooms, executive suites, public debates, and AI itself could tilt even further male,” directly addressing the need for women’s participation in decision-making roles.
   • Target 5.b: Enhance the use of enabling technology… to promote the empowerment of women. The article’s central theme is overcoming the barriers that prevent women from using generative AI, with the goal of turning it into a tool for empowerment, leadership, and “flexible income generation.”
2. Under SDG 8 (Decent Work and Economic Growth):
   • Target 8.5: Achieve full and productive employment and decent work for all… and equal pay for work of equal value. The article connects AI adoption to the gender pay gap, citing a study where AI “advised women to ask for significantly lower salaries than men” and warning that productivity gains could “accrue disproportionately to men.”
   • Target 8.2: Achieve higher levels of economic productivity through… technological upgrading and innovation. The article discusses how generative AI tools “deliver incredible efficiency gains” and are “reshaping how work is done,” which is a direct reference to technological upgrading for productivity.
3. Under SDG 10 (Reduced Inequalities):
   • Target 10.3: Ensure equal opportunity and reduce inequalities of outcome, including by eliminating discriminatory… practices. The article supports this target by advocating for regulatory action like the “AI Civil Rights Act” which would “systematizing bias surveillance in AI systems” to prevent discriminatory outcomes.
   • Target 10.2: Empower and promote the social, economic and political inclusion of all, irrespective of… sex. The article’s call to action is to prevent AI from widening inequalities by ensuring women are not left behind in the adoption and development of this technology, thereby promoting their economic inclusion.
4. Under SDG 4 (Quality Education):
   • Target 4.4: Substantially increase the number of… adults who have relevant skills, including technical… skills, for employment, decent jobs and entrepreneurship. The article emphasizes the need for women to gain AI competence through training to maintain their career advantages, access promotion opportunities, and create new businesses.

3. Are there any indicators mentioned or implied in the article that can be used to measure progress towards the identified targets?

1. Gender Gap in AI Usage:
   • The article directly cites a “Harvard Business School meta-analysis… [that] found women had 22 percent lower odds of using generative AI than men.” This percentage serves as a direct, measurable indicator of the gender gap in technology adoption (relevant to SDG 5).
2. Bias in Algorithmic Outputs:
   • The article mentions a study where “generative AI chatbots advised women to ask for significantly lower salaries than men with identical profiles.” Auditing AI systems for such biased outputs can serve as an indicator of progress towards reducing algorithmic discrimination (relevant to SDG 8 and SDG 10).
3. Productivity Gains:
   • The article references a study by The Federal Reserve Bank of St. Louis which “found that the majority of respondents who used generative AI tools in the past week saved between 2-4 hours.” This time-saving metric is a clear indicator of productivity improvements (relevant to SDG 8).
4. Gender Disparity in Job Disruption:
   • The article cites a prediction that “women are nearly three times more likely than men to be in jobs that generative AI can easily automate.” Tracking the actual rates of job displacement by gender as AI is adopted would be a critical indicator of its impact on inequality (relevant to SDG 8 and SDG 10).
5. Corporate Transparency and Accountability:
   • The article mentions that the proposed AI Civil Rights Act would require “independent audits, algorithmic impact assessments, [and] transparency reports.” The number of companies publishing such reports would be an indicator of progress towards institutional accountability (relevant to SDG 16).

4. Table of SDGs, Targets, and Indicators

Source: ssir.org