Photo by Mikhail Nilov via Pexels

Inclusive strategies to ensure women’s participation in the AI ecosystem are urgently needed to avoid further exacerbating gender inequalities, according to the Asian Development Bank (ADB).

In a report published last month, the ADB highlights that inclusive AI design, deployment and governance could reduce inequalities, expand women’s economic opportunities, improve access to services and enhance safety in digital spaces.

Yet “structural barriers and ‘gender-blind’ national AI strategies limit women’s digital participation and leadership,” it says.

AI is rapidly reshaping economies and societies, with global revenues projected to hit US$407bn by 2027, but the report notes that women hold only 30% of AI professional roles globally and 8% of senior technical roles in Southeast Asia.

To harness AI for women’s empowerment, it recommends integrating gender equality in AI policies and oversight; codesigning initiatives with gender-inclusive tools; expanding funding for women-led AI solutions; and strengthening women’s digital literacy, STEM participation and leadership pipelines.

“AI systems are shaped by the data, design choices, and assumptions of their creators. Without inclusive approaches from the outset, these systems risk perpetuating bias and creating harm,” it says.

Read more: Women’s Network news round-up: From gender bias in AI to Australia’s equality strategy

Inclusive governance and policy reform

The report says governments and public institutions play a “critical role” in developing inclusive AI ecosystems through integrating gender considerations across national AI strategies, policy and regulatory frameworks.

The study highlights that most national AI strategies in Asia and the Pacific lack explicit consideration of gender equality.

As of 2023, 14 Asian countries included in Oxford Insights’ Government AI Readiness Index had released national AI-related policies, such as strategies, roadmaps or action plans, “indicating increasing institutionalisation of AI governance in the region”.

However, only two of these countries – Malaysia and Pakistan – “included explicit references, targets or initiatives related to gender equality or women’s empowerment, underscoring a significant gap in inclusive policy design”.

Six other countries’ policies acknowledge gender, “but without prioritising it or translating recognition into concrete actions”, while the rest “refer only indirectly to inclusion through general non-discrimination provisions”.

Therefore, one of the ADB’s four recommendations is for governments to embed gender equality into AI policies and oversight mechanisms, including the use of sex-disaggregated and localised data as well as gender audits and bias assessments for public AI systems.

Operationalisation is “critical”, it says, stressing that inclusive governance must be supported by resource allocations, enforcement mechanisms, and institutional capacity “to ensure that gender-responsive policies are implemented in practice”.

It also recommends that women’s rights organisations be engaged as cocreators in regulatory design and not just consulted with as stakeholders.

“Inclusive AI governance requires performance metrics that go beyond accuracy or efficiency to assess inclusion, agency and wellbeing. Without gender-responsive governance, risks such as TFGBV [technology-facilitated gender-based violence], data discrimination, and algorithmic exclusion are likely to persist, undermining the transformative potential of AI for inclusive and equitable development,” it says.  

Read more: Baltic states tackle gender bias in AI

Support gender-inclusive tech innovation

Other recommendations include mainstreaming gender in AI design and use; supporting gender-inclusive technology through finance and innovation; and closing capacity gaps.

On mainstreaming gender in AI design and use, the report says women face barriers in accessing and safely using AI-enabled tools “which often limits their role to passive users rather than active participants”.

It recommends that development programmes integrate gender considerations at the outset and that tools be designed that work in low-literacy and low-connectivity environments while “embedding feedback loops to detect and respond to risks”.

To unlock the potential of women entrepreneurs and innovators, it urges public investors to establish dedicated financing for inclusive AI, noting that when women-focused AI innovations are adequately resourced, “they can scale rapidly and deliver both economic and social impact”.  

By directing resources towards applications that address critical gaps, such as early warning systems, maternal health, education, or digital safety, “finance can serve as a powerful lever for shifting the AI ecosystem toward inclusion and equity”, it adds.

On closing capacity gaps, the report notes that underrepresentation of women in STEM, AI research and digital leadership continues to reinforce gender bias in technology systems.

It says capacity-building efforts should expand AI and digital literacy among women and girls, “with a strong focus on education-to-employment pathways, community-based learning, and vocational training in underserved and vulnerable areas”, and that training should extend to developers, engineers and regulators to equip them with the tools to detect and mitigate bias in AI design and deployment.

“Investments in women’s leadership are also essential, ensuring that diverse perspectives are represented in academia, industry and governance bodies,” it adds. “By addressing both foundational digital skills and leadership pathways, capacity development can create a pipeline of women who are not only users, but also shapers of AI.”

The report concludes that “with inclusive design, equitable access, and responsive governance, AI can drive not only technological advancement, but also social and economic empowerment”.   

Read more: G7 AI challenge launches to transform public services

Sign up: The Global Government Forum newsletter provides the latest news, interviews and features on AI, data, workforce, and sustainability in government



Source link