AI and women at work: 5 positives, 8 negatives

The ongoing debate surrounding artificial intelligence and work typically concentrates on the jobs that will be eliminated, the tasks that will be automated, and the skills and qualities that will be most valued in the future. However, fewer articles delve into the impact of AI on exhaustion and pleasure in the workplace, often treating the experiences of men and women as equivalent.

The reality is that women and men are concentrated in different occupations, perform distinct tasks within organizations, and are evaluated differently for the same work, leading AI to affect women's work in specific ways. The overall effect of AI on women's work is neither definitively positive nor negative, prompting this analysis of five potential positives and eight potential negatives.

On the hopeful side, first, today's job growth is predominantly occurring in sectors traditionally dominated by women, such as healthcare and social assistance. In the United States, these sectors accounted for nearly all net job creation in 2025, adding nearly 400,000 healthcare jobs and 300,000 social assistance jobs. While LLMs (large language models) cannot replace caregivers, they do protect women's employment in these crucial fields.

However, care work continues to be poorly compensated, and its growth is dependent on political decisions made in upcoming midterms.

Secondly, AI has the potential to make "greedy jobs" less hostile to women. Many high-paying occupations are characterized by a focus on presenteeism and excessive workloads, which disadvantage those who cannot commit full-time effort. If AI enables the easy access to clients' or case histories, it could reduce the premium placed on being irreplaceable.

This shift could be reminiscent of the impact of digital records on pharmacy, which transformed from a male-dominated, unequal profession to a more equitable one in the US. Nevertheless, employers must decide to redesign these jobs, and this outcome is not guaranteed.

Thirdly, some of the most tedious office tasks are currently being automated, offering relief to women who often bear the burden of these non-promotable responsibilities. Writing meeting minutes, taking notes, organizing schedules, and drafting follow-up emails fall predominantly on women's shoulders. AI tools that automate these tasks free up time that could otherwise be spent on more visible tasks that could lead to promotions.

Additionally, AI can reduce the paperwork in caregiving professions, allowing nurses and doctors to focus more on patient care, which they find most rewarding.

Fourth, AI lowers the barriers to entrepreneurship and technology development for women. Historically, women who started businesses had to work with limited resources. In the US, female-founded startups only secure about 2% of venture capital funding, and this share has remained stagnant. AI allows entrepreneurs to launch small ventures and scale them effectively using "agents."

Women can create websites, draft contracts, and interact with customers using minimal investment. Furthermore, AI-assisted coding enables individuals to develop digital tools without extensive coding knowledge, breaking down barriers for women in the tech field.

Lastly, relational skills are becoming increasingly valuable as AI handles more codifiable aspects of cognitive work. Women have developed these skills over decades due to their role in caring for others and managing their emotions. As AI reduces the importance of codifiable tasks, the market may start valuing relational skills more highly, potentially leading to a change in how society perceives traditionally feminine qualities as valuable assets.

On the worrisome side, the most exposed white-collar jobs, disproportionately held by women, are particularly vulnerable to AI disruption. In rich countries, 9.6% of women's jobs fall into the highest category of exposure to generative AI, compared to 3.5% of men's jobs. Women are concentrated in clerical occupations such as assistants, secretaries, payroll clerks, and data entry clerks.

While not all these jobs will disappear, many may undergo significant changes. For instance, translators, predominantly women, have experienced significant job loss to generative AI, often replaced by the need to edit AI-generated content for reduced pay.

Secondly, the lower rungs of the career ladder, once a viable route to employment and upward mobility for women without higher education, are at risk of disappearing. Junior administrative and clerical roles can lead to promotions, but AI's automation could eliminate these entry points, cutting the career ladder off at the bottom. This development is particularly detrimental to women, who traditionally relied on these positions to advance their careers.

Thirdly, women who utilize AI may face a competence penalty. In a study involving over 1,000 engineers, identical code was evaluated differently based on the gender identity of the coder. Women who used AI tools were judged less favorably, indicating that relying on technology could be perceived as a weakness in certain professional settings. This phenomenon highlights the potential challenges women may encounter as they integrate AI into their work environments.