Girls in AI: Driving Development and Innovation

April 23, 2026 marks International Girls in ICT (Information and Communication Technologies) Day, an annual observance established by the International Telecommunication Union (ITU) to spotlight the barriers girls face in technology and to celebrate those pushing past them. This year’s theme, "AI for Development: Girls shaping the digital future," comes at a particularly consequential moment.

Today, artificial intelligence sits at the center of nearly every major system shaping our economies, societies, and daily decisions. It is embedded in how governments distribute services, how healthcare gets delivered, how financial systems decide who qualifies for credit, and how students access education. The people who design these systems carry an enormous responsibility, and right now, girls and women are significantly underrepresented among them.

That is not simply a demographic observation. It has direct consequences for the quality of AI systems and the communities they serve.

What the Numbers Reveal About Who Gets Left Out

The scale of the gap becomes clearer when you look at the data across multiple levels. According to UNESCO, Women and girls are 25% less likely than men to use digital technology for basic tasks and four times less likely to know how to program computers, and 13 times less likely to file for an ICT patent. These are not minor disparities. They represent a compounding disadvantage that begins with access, grows through education, and eventually shapes who enters the workforce and who builds the tools the rest of the world uses.

The picture inside the AI industry is equally stark:

• Only 20% of employees in technical roles at major machine learning companies are women.
• Women account for just 12% of AI researchers globally.
• Only 6% of professional software developers are women.

In the least developed countries, only 19% of women are online compared to 86% of men in some regions. When girls cannot reliably get online, they are not just absent from social media. They are excluded from learning platforms, skills training, job opportunities, and the very conversations shaping how AI for development gets designed and deployed.

These numbers make one thing clear: the under-representation of girls and women in AI is a fundamental barrier to building inclusive and effective digital solutions for development.

Why Inclusion in AI Is a Development Imperative

There is a tendency to frame girls in AI as a matter of fairness, which it is, but that framing alone does not capture the full stakes. When the people building AI systems come from a narrow demographic slice, those systems tend to reflect a narrow set of assumptions, needs, and blind spots. UNESCO has drawn attention to this problem in concrete terms, noting that the biases people carry in everyday life can be reflected and amplified through the development and use of AI systems, citing the example of how the gendering of digital assistants reinforces understandings of women as subservient and compliant.

This matters for AI and digital future outcomes because AI is the infrastructure of development itself. Credit scoring models determine whether a woman entrepreneur in a low-income country can access a loan. Agricultural AI tools can help or miss smallholder farmers, many of whom are women. Health diagnosis systems trained on male-dominant datasets can underperform for female patients. When girls are absent from the rooms where these tools are built, the tools end up solving for someone else.

The economic argument runs alongside the equity one. According to the GSMA, closing the gender gap in mobile internet adoption in low and middle income countries could add more than $1.3 trillion in gross domestic product to the global economy. Mobile money services have already shown significant benefits, particularly for women-led households, by improving access to financial resources. Greater participation in AI development could generate returns of similar magnitude, across sectors and geographies.

To see these challenges and solutions in action, Africa provides a compelling example.

Africa as a Case Study in What Becomes Possible

Across Africa, the story of girls in AI is already being written in real time, not in policy documents, but in innovation hubs, coding bootcamps, and community programs that are doing the daily work of inclusion.

AfriLabs, with a network of over 500 innovation hubs across 53 African countries, is playing a critical role in building inclusive ecosystems that actively encourage, train, and support young girls to thrive in the digital space. The outcomes from these programs are not abstract. Fatima, a 17-year-old from northern Nigeria, discovered her passion for artificial intelligence through a local hub affiliated with AfriLabs and is now developing a mobile application that uses machine learning to help farmers detect crop diseases. This is precisely what AI for development looks like when girls are given the tools and support to participate: locally relevant, practically impactful, and designed by someone who understands the problem from the inside.

Thandi from South Africa joined a robotics club as the only girl and now leads it, training the next generation of girls in her community. Her story illustrates something important about how inclusion compounds. When one girl gets access and succeeds, she often becomes the role model who makes the path feel real for others who come after her. Visibility generates momentum in a way that statistics alone cannot.

The Barriers Are Specific and They Require Specific Responses

It would be convenient if the solution to underrepresentation were a single policy or program. The barriers girls face in AI and digital development are layered, and addressing them requires responding to each layer rather than treating inclusion as a checkbox.

The challenges compound across multiple fronts:

• Rural and underserved schools rarely have access to coding labs, robotics kits, or reliable internet, meaning girls outside urban centers encounter these fields only by exception, not by design.
• Social norms in many communities still position technology as primarily male territory, shaping what girls believe they are permitted to pursue before they ever encounter a computer science classroom.
• Affordability constraints and online spaces that too often exclude or endanger women and girls make the act of going online feel risky or financially out of reach for many families.
• The absence of visible women in AI leadership means that even girls who develop genuine interest in the field often cannot picture a clear path from where they are to where they want to go.

Despite rising awareness, only 35% of STEM graduates globally are women, a share that has barely moved in the past ten years. This lack of progress means we are still missing out on the creativity, insight, and lived experiences of millions of talented girls who could be shaping the AI solutions of tomorrow.

The pipeline problem is not self-correcting; it requires active, sustained intervention at multiple points, from early education through career entry and advancement.

Also Read: How Women AI Leaders Turn Risk Awareness Into Technical Excellence

Programs That Are Changing the Equation

Several initiatives are demonstrating what intentional investment in girls in AI actually looks like in practice. The ITU’s EQUALS Global Partnership, now entering a new phase, aims to equip 100 million women and girls with digital skills by 2035, while the AI Skills Coalition brings together tech leaders and academic institutions to offer free, tailored training in artificial intelligence.

At the community level, targeted mentorship programs, community-driven coding bootcamps, and inclusive design practices are opening doors for thousands of young women across the continent, with girls becoming creators, problem-solvers, and entrepreneurs rather than just technology users.

Corporate initiatives are also contributing. Ana Cidre, founder of GalsTech and Head of International Developer Relations at Okta, built her community precisely because the numbers told a clear story: in the UK, women make up 47% of the overall workforce but only 21% of IT specialists. GalsTech was designed to counter this by creating mentorship connections, facilitating peer learning, and showing younger girls that the developer community has space for them. The Women@Okta platform extends this further, with an internal speaker series and structured mentorship to support women at every career stage.

What these programs share is an understanding that access alone is insufficient. A girl who gets online but has no one to guide her, no community to belong to, and no role model to aspire toward is still facing a gap, just a different one. Effective inclusion addresses the full chain from connectivity to confidence to career.

Conclusion

The theme for International Girls in ICT Day 2026 asks something specific of the institutions, companies, and communities that shape how AI gets built and who gets to build it. It is a call for the kind of structural investment that changes what a girl in a rural school or an urban slum believes is possible for her.

That investment looks like government digital education policies that explicitly prioritize gender inclusion from the design stage. It looks like companies that measure not just hiring numbers but retention, promotion, and the degree to which women’s voices actually shape product decisions. It looks like communities that treat a girl’s curiosity about coding as something worth nurturing rather than redirecting. And it looks like AI researchers and practitioners who take seriously the question of whose needs their systems are designed to serve.

The real cost of the current gap is in the untapped potential of girls who could be the coders, entrepreneurs, engineers, and digital diplomacy leaders that the world desperately needs. Girls in AI are not a future investment whose returns will arrive in a decade. They are already building crop disease detection apps, leading robotics clubs, and designing tools that serve their own communities better than any outsider could.