The gender digital divide is the disparity in digital technology access, usage, and skills between women and men, often compounded by intersecting factors such as income, geography, and education. It is not merely a matter of hardware availability but encompasses gaps in connectivity, digital literacy, online participation, and representation in tech fields, including the development of Artificial intelligence systems. These gaps shape who benefits from technological progress and who is left behind, with implications for economic opportunity, social inclusion, and the design of emerging technologies.
Measuring the divide has been a focus of international organizations. The International Telecommunication Union (ITU) has tracked gender-based internet usage statistics since the early 2000s, reporting that in 2023, the global internet usage rate for women was approximately 63 percent, compared to 69 percent for men, leaving an estimated 244 million fewer women online than men. The gap is most pronounced in low- and middle-income countries, particularly in sub-Saharan Africa and South Asia, where mobile phone ownership also lags significantly. Data from the GSMA's annual Mobile Gender Gap Report in 2024 indicated that women in low- and middle-income countries are 15 percent less likely to own a smartphone than men, a figure that has remained stubbornly stable in recent years.
Drivers of the Divide
Socioeconomic and cultural factors are primary drivers. Cost remains a leading barrier, as women often have less disposable income and face higher relative costs for data and devices. Social norms in many regions restrict women's mobility and autonomy, limiting their ability to visit internet cafes or purchase mobile phones. Educational attainment correlates strongly with digital literacy, and gender gaps in schooling, especially in rural areas, hinder women's ability to navigate digital interfaces. Safety concerns also play a role; surveys conducted by the Pew Research Center and others consistently find that women experience higher rates of online harassment, which discourages engagement. In addition, a lack of relevant content and digital skills training tailored to women's needs perpetuates low usage patterns.
The Divide in Artificial Intelligence
Within the technology sector, the gender digital divide extends to the workforce and the design of AI systems. Women hold a minority of technical roles at major AI companies, including OpenAI, Anthropic, Google DeepMind, and Amazon Web Services, with estimates suggesting women make up around 25 percent of AI research positions globally. This underrepresentation shapes the development of Machine learning models. For instance, natural language processing systems trained on biased text corpora can inherit gender stereotypes, leading to algorithms that associate doctors with male pronouns and nurses with female ones. Studies by researchers such as Timnit Gebru and others have documented these biases in Large language model outputs. AI assistants like those used in hiring tools, developed with skewed datasets, have historically penalized resumes containing the word "women's" or graduates from all-female colleges.
The lack of diversity in teams at places like Deep learning labs and Neural network research groups also affects product priorities. Autonomous vehicle development at companies like Waymo and Tesla has been criticized for largely using male crash-test dummies, which may affect safety for female passengers. Similarly, health AI, such as that from Intuitive Surgical, relies on data that sometimes overlooks sex-specific differences in physiology.
Economic and Social Consequences
Women's exclusion from the digital economy has measurable economic impacts. The World Bank has estimated that closing the gender digital divide could add billions to global GDP, as women gain access to online markets, mobile banking, and remote work opportunities. In agriculture, women farmers using mobile advisory services improve yields, but they are less likely to own the necessary devices. The Bhabha Atomic Research Centre and other public institutions have initiated programs to provide rural women with training in digital literacy, but scale remains limited. Educational platforms, including those from MIT CSAIL and Stanford AI Lab, offer open courses, yet women's enrollment in advanced computing programs lags, exacerbating future workforce gaps.
Policy and Industry Responses
Governments and NGOs have launched targeted initiatives to bridge the gap. The United Nations' International Girls in ICT Day (held annually in April) encourages member states to support young women's tech education. Organizations like the GSMA's Connected Women program have trained millions of women in mobile literacy. Private sector efforts include Amazon Web Services' education scholarships and Google Cloud's jumpstart programs for women in cloud computing. Some countries, such as Rwanda and Kenya, have implemented gender-sensitive broadband policies, increasing female internet usage rates by over 10 percentage points between 2019 and 2023. Notably, Arm Holdings and Nokia Bell Labs have sponsored research into low-cost connectivity solutions designed for underserved communities, recognizing that infrastructure alone does not ensure equitable use.
Future Directions
As AI systems become more integrated into daily life, addressing the gender digital divide becomes critical to avoiding automated discrimination. Researchers argue for more inclusive data collection practicesable and transparent algorithm audits awareness of models from OpenAI and other firms. Initiatives at Xerox PARC and BAIR (Berkeley AI Research) are exploring participatory design methods where women co-create technology. The shift towards Generative AI tools adds urgency; if women are not equal users and producers, these tools will likely be designed from a male-centric perspective. Closing the gap requires multi-stakeholder action: improved connectivity in rural areas, lower device costs, culturally sensitive digital literacy training, and systemic changes in hiring and product design. The evidence from regions like South Korea and Scandinavia, where targeted policies have reduced the usage gap to below 2 percent, indicates that progress is possible with sustained commitment.
Measurement and Critiques
Scholars note that the gender digital divide is not a binary but a spectrum, intersecting with age, disability, and rural-urban status. Surveys often undercount informal use, such as shared device access or borrowed phonesaboriginal social media use. Some critics argue that focusing solely on women's deficits overlooks structural inequities, recommending a 'gender-transformative' approach rather than simple inclusion. The OpenPanel coalition, a civil society group, has proposed ethics-based standards for tech companies to report workforce diversity, which several firms, including Intel and AMD, have begun adopting, though disclosures vary. The lack of sex-disaggregated data remains a persistent challenge in tracking progress.