AI was supposed to be the great equalizer: A tool that could help workers move faster, communicate clearly, and compete on the strength of their ideas, not their access to resources. But early signs suggest AI may also become something else: a new arena for old gender biases.
A recent Fortune article highlighted research showing that identical AI-assisted resumes were interpreted differently depending on whether they came from a man or a woman. In the study, two resumes were generated with the same content and the same use of AI. The only meaningful difference was the name at the top: Emily Clarke or James Clarke. Reviewers of Emily’s resume were more likely to question whether she could be trusted, and her resume was more likely to raise doubts about competence and ability. James’s AI use, by contrast, was more often excused or interpreted favorably.
Access alone will not solve this. Women may have the same AI tools available to them, but if they are less likely to be encouraged, trusted, or rewarded for using them, the gap will only grow.
For years, women in the workplace have navigated conflicting expectations. Be confident, but not arrogant. Be collaborative, but still prove individual competence. Be efficient, but do not appear to cut corners. Be innovative, but do not seem risky.
The data is already pointing in the wrong direction. Studies confirm that gender gaps in generative AI use are nearly universal across regions and sectors. This gap is already manifesting: men are 27% more likely to be praised for using AI on the job than women are.
This is not an isolated hiring issue, it is a pervasive workplace issue.
It is also not only about whether women know how to use AI, but whether they feel they have permission to use it without being judged differently.
When a man uses AI to sharpen a memo or prepare for a presentation, he may be seen as efficient or strategic. When a woman does the same, is she seen as less capable, relying on a crutch, or taking an inappropriate shortcut? These are not theoretical questions.
The risk is not that AI itself becomes the glass ceiling. The risk is that organizations adopt AI on top of old credibility gaps and then mistake unequal outcomes for individual choice.
If women believe they will be penalized for using a tool that men are praised for using, many will opt out. Not because they lack curiosity, ambition, or technical ability, but because they understand that the same behavior does not always carry the same consequences.
This has real implications for advancement.
AI fluency is quickly becoming a career advantage. Employees who learn to use these tools well may move faster, produce more, identify new opportunities, and help shape how their organizations adapt.
If men are encouraged to experiment and women are quietly discouraged—or if men are rewarded while women are scrutinized—the AI gap will become a promotion gap, a pay gap, a leadership gap, and eventually a power gap.
That is how glass ceilings are built. Not always through explicit exclusion, but through accumulated differences in trust, opportunity, feedback, and recognition.
This outcome is not inevitable.
Leaders must treat AI adoption as a culture challenge, not just a technology rollout.
That starts with clear norms. Organizations need to define what responsible AI use looks like. Where is AI encouraged? Where should it be disclosed? What requires human review? What is unacceptable?
It also requires equitable manager training and performance review standards. Managers are often the ones who decide whether AI use is seen as initiative or inadequacy, so organizations must look honestly at how AI use is evaluated. Without shared standards and objective judgment, employees receive inconsistent signals; women must receive the same encouragement, training opportunities, and public recognition as their male counterparts when testing tools and building AI-enabled workflows.
If two employees use the same tool for similar work, are they judged the same way? Are women receiving the same encouragement to test tools, attend training, lead pilots, and build AI-enabled workflows? Are they receiving the same public recognition when those efforts succeed?
Organizations must stop treating AI literacy as informal or optional; instead, AI training should be structured, practical, and available across all roles and levels, ensuring it is not reserved only for technical teams or early adopters.
The true competitive edge comes not from buying new tools, but from building a culture where people can use them with confidence, clarity, and fairness.
That includes women.
Because if the future of work is being shaped now, women cannot afford to enter it already carrying a credibility penalty.
AI may be new, but the double standard is not. But leaders have a chance, right now, to interrupt the pattern before it becomes embedded in how work is evaluated.
They can set clear expectations. They can train managers. They can reward responsible experimentation. They can track who is encouraged, who is praised, who is promoted, and who is left behind.
Most of all, they can be honest: Technology alone will not fix culture. If anything, AI will reveal it.
And if what it reveals is that men are praised for using new tools while women are questioned for the same behavior, then the problem is not the tool.
The problem is the double standard we allow into the future of work.
Ready to FLEX?
When you're ready for strategic support that adapts to your unique needs, FLEX Partners is here to help. Connect with us to explore how we can empower your success.