Opinion by Dr Fauzia I.Abro
AI is no longer a futuristic promise but an everyday reality shaping how we learn, work, and connect. Its potential to expand opportunities and create safer, more inclusive societies is immense. Yet, without responsible governance frameworks, AI risks widening inequality, increasing discrimination, and eroding trust in digital systems. For women and girls who are already underrepresented in technology and exposed to online harm, the stakes are particularly high.
The challenges women face in digital environments are now amplified by AI. Harassment, cyberbullying, image-based abuse, and exclusion from digital economies persist across platforms. Evidence already shows that AI, when designed on biased data or deployed without oversight, can magnify these harms. The Gender Shades study by Buolamwini and Gebru (2018) revealed that commercial computer vision systems misclassified darker-skinned women at error rates many times higher than lighter-skinned men, a stark reminder that biased data yields biased outcomes. This is not surprising when women represent only about 22 percent of the global AI workforce (ITU, 2024). Their limited presence in design teams and governance structures means their perspectives and experiences are often absent from the systems shaping digital futures.
These exclusions have had harmful consequences. In India, AI-driven healthcare pilots misdiagnosed women because female patients were underrepresented in training data. In Australia, the Robodebt scheme, an automated welfare debt recovery programme, disproportionately harmed single mothers and vulnerable families, demonstrating how systems designed with biased algorithms can deepen inequalities. Such failures are compounded by the fact that AI systems are often deployed without transparency or accountability. The result is erosion of trust not only in technologies but also in the institutions that oversee them. If women and girls cannot trust AI to treat them fairly, they are less likely to participate fully in digital societies.
AI does, however, present opportunities to reimagine digital environments as safer and more inclusive if designed and deployed responsibly. Machine learning can detect patterns of online abuse, enabling interventions before harm escalates. AI can also support women’s access to education, health, and financial services, particularly in regions where opportunities are otherwise constrained. Yet whether
these possibilities are realised depends on how AI is governed. Innovation designed solely around economic growth and commercial gain sidelines fairness, transparency, accountability, and inclusion.
For women and girls, this risks exclusion from both the design of AI systems and the benefits they promise.
WISE governance—Women-Inclusive, Safe, and Ethical governance—offers a way forward. It begins
by embedding inclusion from the outset. Diversity cannot be left to voluntary initiatives; it must be
present at all stages of the AI lifecycle and reinforced through national AI strategies, procurement rules,
funding priorities, and leadership pipelines that require gender parity and representation of marginalised
voices. AI systems cannot be truly safe or ethical if women are absent from design, testing,
implementation, governance, and policymaking.
Safety, the second pillar of WISE governance, must be measured in lived experience as well as algorithmic performance. While AI can help flag harassment or detect image-based abuse, governance should mandate transparency in moderation algorithms, regular publication of error rates disaggregated by gender, and robust user rights to appeal wrongful decisions. Without such protections, AI risks becoming yet another layer of control over women’s digital lives.
Ethics, the third pillar, demands moving beyond corporate codes of conduct to enforceable frameworks.
UNESCO’s Recommendation on the Ethics of Artificial Intelligence (2021), places human dignity, non-
discrimination, and accountability at the centre. These frameworks, however, risk remaining symbolic
unless translated into binding obligations. National regulators must require gender impact assessments
for high-risk systems, mandate independent audits, and establish clear avenues for redress when harms
occur.
The GSMA Mobile Gender Gap Report (2024) highlights that in low- and middle-income countries,
women are 15 percent less likely than men to use mobile internet. AI cannot empower those disconnected from the infrastructure that delivers its services. Governance frameworks must therefore
integrate AI policy with investments in connectivity, affordability, and digital-skills programmes designed for women and girls. Empowerment comes not only from being protected by governance but also from being able to shape, contest, and direct AI systems themselves.
The test for WISE governance is simple: does it expand women’s and girls’ ability to learn, work, participate, and feel safe online? If the answer is no, then the system is neither safe nor ethical. Meeting
this test requires intentional inclusion of women in every stage of AI design, governance, and oversight.
It requires investment in digital infrastructure and skills to close the gender gap in access. And it requires
enforceable ethical standards that ensure transparency, fairness, and recourse when harms occur.
Responsible AI governance—Inclusive by Design—can transform AI from a mirror of old inequities into a tool of empowerment. The choice is not between innovation and inclusion, but whether we have the will to govern wisely.