Internet Explorer 11 is not supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

AI's Impact on Roles Held by Women

In organizations where AI has the potential to significantly impact women's roles, women can offer the deepest understanding of how it can enhance their work.

GOV KPMG Special Issue_Her
iStock
Editor's Note: This article appears in Governing's Special Issue 2025 magazine. You can subscribe here.

As artificial intelligence becomes increasingly prevalent in the public sector, it will impact some agencies and roles more than others.

Generative AI (GenAI) tools are most likely to disrupt “cognitive” roles that require advanced degrees, as well as administrative support positions, according to a Brookings Institution report. In fact, more than 70% of office and administrative functions could be automated by GenAI, the report found. Another study, by IBM, found that GenAI is being used most commonly in customer service, human resources and marketing departments.

There’s a common thread that runs between all those trends: These roles and offices are typically held by women.

“Women are more likely to be in jobs that could be substituted by generative AI,” says Molly Kinder, a Brookings Institution fellow and one of the authors of the GenAI report. “But the technology can also be a superpower for people to do their tasks faster, to be more creative, to learn new skills and to get incredible feedback.”

Despite the potential for AI to enhance the work experience, many women across both public and private sectors have concerns about AI that go beyond those of their colleagues. While women in senior technical roles are equally or even more likely than men to embrace AI tools, women in non-technical roles at all levels are less likely than men to adopt and embrace the technology, according to a global Boston Consulting Group survey. Women are also more worried about negative workplace implications: According to the IBM study, women are more concerned than men about their roles being eradicated by AI.

AI also poses the risk of exacerbating existing gender gaps and other inequities in the workplace due to algorithmic bias. “If a tool is trained on biased data, it will perpetuate existing issues,” says Amy Hille Glasscock, program director of innovation and emerging issues for the National Association of State Chief Information Officers. “If you use an AI tool for hiring or performance evaluation, the algorithm could favor, for example, male candidates if men had been dominant in those roles historically.”

GOV KPMG Special Issue_her3
Adobe Stock

The reality is that while women will be more impacted by AI, they’re also less likely to have a say in how it’s implemented. That’s a big problem, says Kinder.

“Everywhere along the chain of generative AI, women are underrepresented,” she says. “But it’s women who are in the bullseye of generative AI. They are the ones who are going to have the most to gain or lose from this technology at work.”

The news isn’t all bad. AI poses unique opportunities for workers, and for women in particular. The same automation capabilities that some employees find threatening could be used to enhance daily work. GenAI can be useful in summarizing documents, writing emails and other materials, retrieving and synthesizing information, and even analyzing data to inform decision-making.

These are tasks carried out in many female-dominated fields. By embracing AI — and with access to proper training and upskilling — women could offload those tasks and focus instead on more creative endeavors that enhance their own work experience.

“Generative AI can be a tool for excelling in your work, saving you time on things you don’t necessarily enjoy doing, and simply being better at your job,” Kinder says.

Enhancing female employees’ experience can have an outsized impact on constituent outcomes, says Lynn Overmann, executive director of the Beeck Center for Social Impact + Innovation at Georgetown University. “Women disproportionately serve as frontline service providers in government,” she says. “When used well, AI can make their jobs easier and more effective.”

Consider benefits administration roles, which require deep knowledge of complicated eligibility policies. “Because many benefits are federally funded and state administered, some states spend years training caseworkers to understand how to navigate all the policy regulations,” Overmann says. “AI can help staff quickly parse through policy documents to get the information they need to make determinations quickly and accurately.” The result is an improved experience for both staff and the public.

When used to ease employees’ workloads, AI can also reduce burnout, which has worsened significantly for women leaders in the workplace since COVID, according to research from McKinsey & Company and LeanIn.org. Women in their research have consistently identified a manageable, sustainable workload as a critical part of professional wellbeing. AI tools have unprecedented power to make that a reality.

Because of AI’s disproportionate impact on female workers, it is imperative for women to be leaders in the rollout of any AI technology, experts say.

In agencies where women’s roles are most affected by AI, women could provide the most insight about how AI can support their work, according to Debra D’Agostino, managing director of thought leadership for Oxford Economics, a global economic forecasting firm.
GOV KPMG Special Issue
Adobe Stock
“You have an opportunity for women to be the first line of offense for AI,” D’Agostino says. “They can identify clear opportunities for their businesses to take advantage of AI or large language models. They can think not just about how to automate work, but how to accomplish things their offices could not do before. There’s a natural opportunity for women in their organizations to step forward.”

This is especially critical in the public sector, where the use of AI can impact not only workers but also the general public. Frontline workers have a keen perspective on how AI could improve services. The person most qualified to help create an AI tool for social services benefits delivery is a caseworker. In K-12 education, the person best suited to evaluate the efficacy of GenAI for creating lesson plans is a classroom teacher. The strategic vision and feedback of these workers is vital to the successful implementation of AI technology.

“Experts in the space — the people who know the constituents — should be part of the decisions around AI,” Kinder says. “This will ultimately be better for the public.”

Women also play a vital role in mitigating the risks of algorithmic bias in new AI technologies, adds D’Agostino.

“AI models need to be trained,” she says. “Leaders need to consider: Are they training them in a way that will consider unintended consequences and minimize risk?”

Successful implementation and adoption of AI tools also require the input of diverse voices who represent the unique needs of both staff and the public. Women whose roles are directly public facing can be particularly valuable for this kind of work.

“Women in the public sector can help strike a balance between serving the public effectively, understanding the technology fully, and identifying the different ways the technology can get deployed successfully or not,” says Kinder. “This is a real opportunity for women.”

It’s important for women to remember that being a leader in AI does not require a STEM degree or even extensive knowledge of AI technology.

“You don’t have to be an expert on AI to figure out how to leverage it for your organization,” D’Agostino says. “What you need to know is your organization. Then you can look at your team and say, ‘Here are the places where AI can help us.’”

Kinder agrees, and encourages women to adopt a curiosity about AI and a willingness to learn more.

“Get smart about these technologies,” says Kinder. “Raise your hand if there’s an opportunity to get involved. When it comes to AI, do what you can to be in the driver’s seat.”
Emma Newcombe has a Ph.D. in American and New England Studies from Boston University.