Artificial intelligence is reshaping work faster than policy or leadership can adapt. U.S. companies report record productivity, yet payrolls barely rise. Goldman Sachs estimates that AI automation could affect the equivalent of 300 million full-time jobs worldwide. Investors are cheering the efficiency. But history suggests that when work becomes scarce, societies ration opportunity, and women often pay the price.

The pattern is familiar. During the Great Depression, dozens of U.S. states and school districts enacted “marriage bars,” policies that barred married women from employment or forced them to resign upon marriage, claiming to “protect” male breadwinners. After World War II, governments closed wartime child-care centers and urged women out of factories so returning soldiers could reclaim work. In post-war Japan and Australia, the “male breadwinner compact” guaranteed men lifetime jobs while women were steered into part time work or unpaid care. Each policy was framed as moral restoration; each was economic triage.

AI may now drive a similar re-ordering. “Headcount-light” companies can scale output without adding workers. Knowledge-based roles once thought immune to automation: legal research, accounting, customer service and the like, are being rewritten by software. For many displaced workers, especially mid-career professionals, retraining programs rarely keep pace with technology’s curve.

As the labor market polarizes, some voices are recasting gender equity itself as a problem. A recent essay by commentator Helen Andrews titled *“Overcoming the Feminization of Culture,” *has drawn unusual attention. Andrews argues that the growing presence of women in professional and public life has made society “empathic rather than rational” and “risk-averse rather than competitive,” and that this “feminization” represents a potential threat to civilization itself. According to The New York Times, as of October 23, her speech had been viewed more than 175,000 times. Her argument resonates precisely because economic anxiety seeks moral explanation. History shows that when structural change threatens status, nostalgia for hierarchy often masquerades as rational analysis.

An economic paradox

The economic paradox is clear. In the short term, investors may reward companies that grow without hiring. But long-term prosperity depends on broad participation in income and consumption. According to the International Monetary Fund, raising women’s labor force participation to men’s levels could expand GDP by up to 35% in some economies. Conversely, excluding women, or any large group of workers, shrinks markets, innovation, and resilience.

Governments under fiscal strain are simultaneously cutting social supports such as child care subsidies and workforce training. If job losses accelerate, the temptation to frame gender regression as cultural renewal will rise. But excluding women from paid work doesn’t just shrink the labor force, it also makes it older.

In most advanced economies, women now supply the bulk of new labor-force entrants in the 25-to-54 age group, the very cohort that offsets aging among men. When women step back or are pushed out, the pipeline of prime-age talent contracts even as older men delay retirement. The result is a workforce that is smaller, less dynamic, and aging faster, precisely when adaptation to technological change requires the opposite.

For boards and investors, this is not a social-policy sidebar; it is a core governance issue. Directors should press management to quantify how AI will change headcount, skill mix, and pay equity over the next five years. They should examine whether algorithmic HR tools introduce hidden bias or legal exposure and ensure that human-capital disclosures explain how automation affects opportunity by gender and age. Insurers and lenders are already incorporating these factors into risk models.

The larger question is one of social license. Companies cannot thrive indefinitely in economies that cannot sustain full employment. A short-term efficiency story can quickly become a long-term demand problem, and, if gender backlash gains political traction, a reputational one.

When societies fear obsolescence, they often seek order through exclusion. The impulse is as old as industrialization itself: when technology or globalization threatens the familiar, institutions reassert hierarchy to restore a sense of stability. Schools once pushed girls out of science when jobs were scarce; factories barred women from higher-paying trades to protect male employment; companies in the 1980s celebrated “decisive” and “tough” leadership as automation hollowed out middle management. Each response framed exclusion as virtue: efficiency, morality, or merit, but all served the same purpose: to make uncertainty feel orderly.

Thus we have seen it before, in classrooms, factories, and corporate hierarchies. The technology has changed; the instinct has not.

AI will redefine how humans create value. Whether it also redefines who is *allowed *to create value will depend on the choices leaders make now.

Efficiency can make a company stronger and a society brittle at the same time. What we choose to optimize will tell us what kind of future we deserve.

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.