While there are many fears about job loss during the rise of AI, women might lose the most. However, if women increased their presence in tech, they could literally write the code to navigate the numerous threats to their job security. These are not unreachable goals, and they are certainly not new. Women have been fighting uphill battles throughout history, from struggling for citizenship or gaining the right to vote to stepping over, around, and through barriers—real or imagined—to keep them from advancing in their careers.
Right now, traditional barriers of gender and cultural bias are being magnified. In the US, the tech itself has reportedly made assumptions about and toward women from mostly male perspectives. The Women in Tech Network reports, women represent less than one-third of the STEM workforce. The report lists additional hits to representation, including women experiencing more layoffs than men (45% more in 2022), and applying at far fewer rates, (only 25% of entry-level applicants).
So what does that do to AI, which is learning about and from us by mostly male-created code? In Michael Connelly’s new novel, The Proving Ground, Mickey Haller (Lincoln Lawyer) argues a civil case where a teenage boy develops a relationship with an AI chatbot who urges him to “get rid” of his human girlfriend, which he does. In the process of holding the tech company responsible for the girl’s murder, Haller found (spoiler) one of the male coders (there were no females) to be a member of a group of men who hated women and transferred his misogynistic attitudes into the code. While this is obviously a fictional worst-case scenario, there are plenty of real-life incidents that demonstrate the ways women are at risk from ongoing bias, and from being left out of the process.
“The data problem is structural, not incidental,” says Shubhi Rao, founder and CEO of Uplevel, creator of UpGenie AI. He reported that, “A 2023 analysis by the Berkeley Haas Center for Equity, Gender, and Leadership examined 133 AI systems and found that roughly 44% exhibited measurable gender bias. The culprit is not bad intentions; it is bad inputs. When the datasets used to train AI over-represent male behaviors, male speech patterns, and male-coded workflows, the intelligence those systems produce is, by definition, intelligence built for someone else.” Finally, Rao stated the overall problem we must solve to bring women as full participants into the age of AI. “AI cannot work for women if gender data doesn’t exist. And no one was building it.”
Women are leaving themselves out of the AI revolution as well. Lean In conducted a study in 2025 to learn how men and women felt about using AI. Overall, men were more positive in all ways. Women were more cautious and even reluctant about using it for ethical reasons. Once they did use it, to assist with coding for example, women were judged less positively and even to be less competent than men. In addition, women were less encouraged by management to use AI than men, and women were more likely to distrust the accuracy of AI’s responses.
Because of these results, Lean In is working to help women understand the importance of becoming full participants in AI. In their newsletter, The Lead, Bridget Griswold, CEO, predicts, “We are at the beginning of a true revolution—one that is profoundly changing the way we live and work…Very quickly, AI will touch nearly every industry and job, in a transformation akin to the Industrial Revolution and the rise of the internet—but faster.” Griswold goes on to encourage women who are not using AI to familiarize themselves with AI tools that might increase productivity in their jobs. She denies that it requires technical skills, and says, “It’s about experimentation and the ability to learn.” She also promises more to come. You can sign up for The Lead here.
It’s also time to end the gender bias. The more women stand on the sidelines and allow it to be built into AI, the more women will continue to suffer some of the harms that are already apparent—like when AI is tasked with reviewing resumés for job applicants. White male names were favored over women of all races and Black men. This was discovered in a Washington University study, using three AI systems that looked at 500 resumés and job descriptions allowing for more than 3 million comparisons. Payal Dhar, reporting for Science News Explores, said that the study was conducted to test claims that AI screening was more fair since resumés provide only data. This is especially significant when you read the statistics from World Economic Forum that “Approximately 88% of companies already use some form of AI for initial candidate screening.”
Voices advocating for more diversity and greater representation can be found from multiple sources. In the article, “Top 10: Women in AI,” published in AI Magazine, Debbie Weinstein who serves as President of Google in the EMEA region, overseeing AI strategy, innovation and market expansion, advocates for diversity in AI leadership and actively supports initiatives aimed at increasing female representation in the tech industry. And she is not alone, check out the interviews with these amazing top 10 women for more inspiration about the future of AI and the roles women are playing in leading the way.
While this is inspirational, more talent is needed. UpLevel warns, “Women are socialized to wait until mastery feels complete before speaking. AI is evolving too fast for that instinct to serve you. Authority now belongs to those willing to think in public, in real time.” They predict that the next 24 months are important to career advancement in AI. When something is this new, it provides motivated people with opportunities to become experts. UpLevel says that men in tech are filling that role. Women need to step up to become architects and help design the future of AI. The more diverse the input, the more valuable the new intelligence will be. Women need to lead, not follow, into this new age, and guide the change toward a future that benefits us all.