AI Is Replacing More Of Women’s Jobs. Can We Build Inclusive Workplaces?
Data suggests employed women are almost twice as likely to work in jobs at high risk of automation. At stake are 65 million jobs for women versus 51 million for men (Photo by Andres Siimon/Unsplash)
The world is now at a critical crossroads: on one hand, there’s the transformative potential of a technological revolution, and on the other, the risk of reinforcing gender prejudices within the digital infrastructure.
For decades, women have faced numerous challenges, including hunger, poverty, domestic violence, the burden of unpaid care work, and child marriage.
Now, there’s also a growing concern of job losses driven by the AI revolution.
In the backdrop of the International Women’s Day, we analyse how the future of AI could be shaped to protect women, and advance a more equitable future.
Algorithmic biases
Generative AI is most likely to disrupt the jobs dominated by women. More women work in office and administrative jobs that are typically exposed to automation, and hence bound to face higher disruptions. This shift could then significantly reshape career pathways and economic stability for many women in the workforce.
According to the The Gender Snapshot 2025 report, which provides comprehensive, data-driven assessment of global progress toward gender equality, especially in relation to the Sustainable Development Goals (SDGs), employed women are almost twice as likely to work in jobs at high risk of automation — 4.7 percent women as compared to 2.4 percent men, representing about 65 million jobs for women versus 51 million for men.
This disparity becomes even more pronounced in high-income countries such as Australia and New Zealand, where 9.6 percent of women’s jobs are at risk compared to 3.5 percent that of the men, as per the report.
Another estimate echoes similar concerns — in the UK, about 119,000 clerical roles in tech and financial and professional service sectors will be automated over the next decade.
Gender biases run deep. Current AI frameworks — largely developed by men — tend to marginalise women. This is because artificial intelligence operates on data and algorithms, and humans shaping it are themselves often steeped in societal biases. Programmers who are training AIs might unconsciously interpret reality through a predominantly male lens.
A 2024 UN report highlights an instance of gender bias in generative AI. While researching a novel, a user prompted the AI to write a story about a doctor and a nurse. The AI automatically assigned the doctor as male and the nurse as female, reflecting its tendency to reinforce gender-stereotypes.
This is particularly concerning as digitisation is constructing an entirely new societal framework built upon data and algorithms that may already carry embedded biases.
A 2018 study by Professor Anja Lambrecht of London Business School and Catherine Tucker of MIT found that job advertisements in STEM (Science Technology Engineering and Mathematics) were shown more frequently to men than women, even when advertisers aimed for equal exposure. The advertisement was shown in 191 countries to men and women over the age of 18.
AI is increasingly being integrated into workplaces to streamline administrative tasks and decision making in recruitment and employee training mapping.
In the modern world of recruitment, algorithmic gatekeepers are increasingly determining who gets opportunities, given that they scan applications much before a human reviewer does. If AI is trained on data that links women and men to specific skill sets, interests, or capabilities – for example, women are better equipped to handle roles that require emotional labour – it will inevitably produce content that mirrors those biases.
Higher representation, balanced data
To eliminate this bias, it is imperative that steps be taken to achieve gender equality in this domain too. First, this requires that we increase the number of women in AI research and the development of technology, so that these fields benefit from a balanced perspective.
Even as we actively challenge all forms of gender bias in AI development, greater representation of women in STEM leadership is essential, and must be encouraged. According to the 2025 World Economic Forum report, women constitute only 28 percent of the global STEM workforce, and just 22 percent of AI professionals.
At its core, the transition to an AI-driven economy will be fundamentally shaped by digital literacy. But the existing digital divide, particularly in the countries of the Global South such as India, further deepens systemic biases. In low-income countries, only 20 per cent of the women have access to the internet, making even basic connectivity a luxury for many women.
The gender gap in smartphone ownership prevents a large number of women from accessing equal opportunities, and competing on a level playing field.
Without equitable access to technology, infrastructure, and quality education, large segments of the population risk being excluded from emerging economic opportunities.
In addition to closing this digital literacy gap, prioritising training and professional development for women workers is essential too, so as to ensure equitable participation in the AI-driven economy.
This proactive approach not only empowers women individually, but it also strengthens overall workforce resilience in an increasingly automated world.
Achieving gender equality in AI is essential to build a fair and inclusive digital future. By empowering women through leadership, education and equitable access, as well as designing AI systems with unbiased data and inclusive frameworks, we can ensure that the AI revolution is not gender blind and benefits everyone.
Because women should not be left at the sidelines of the digital revolution.
