Women and AI: 4 questions to ask about opportunity and power in the AI economy

From the early days of the artificial intelligence (AI) transition, women have shown lower levels of engagement with AI than men. This adoption gap has been too easily framed as a participation problem and its solution, that women need to catch up, too quickly accepted.

In that process, we have missed out on the opportunity to understand what the gap itself reveals about how the AI economy is being built.

Across the AI ecosystem, there are spaces where parity and opportunity are being priced, allocated and distributed and where women’s ambivalence points to kinks in the chain of how AI is adopted, developed and invested in and for whom.

Here are four questions that could lead us to better gender parity in AI.

Much of the conversation around AI and gender has centred on women’s exposure to automation and displacement, with justified concern.

Early research from the World Economic Forum found that women workers were disproportionately represented in occupations exposed to technological disruption, yet underrepresented in sectors expected to benefit most from AI-driven productivity gains.

And even within the AI workforce, women were estimated to be more likely than men to occupy lower-status, lower-paid roles in the data and AI talent pool, despite having, on average, better credentials, thereby limiting their access to the high-value gains generated by the AI economy.

Yet, focusing exclusively on who might lose out risks overlooking the question about who might gain and where. AI does not spread evenly across the economy. It tends to flow toward sectors where gains are easiest to measure, scale and monetize.

The Stanford AI Index Report shows that AI-related job postings remain highly concentrated in information, professional, scientific and technical services, as well as in finance and insurance – all sectors with high-male workforce representation.

The picture is different in, for example, in healthcare, a sector whose workforce is predominantly female, and where demand for care workers has only increased in recent years but AI jobs are not yet distinctively present. That does not mean care jobs would not benefit from AI augmentation to reduce labour-intensive tasks such as documentation, scheduling, reporting and information management.

Augmentation, implemented with consultation, training and labour protections, could well improve institutional trust, service quality and workplace health and safety, reduce staff burnout and even yield long-term public expenditure savings.

Mixed results have emerged from initial pilots in economies such as Japan. However, a broad-based appetite for AI interventions in these sectors remains less pronounced.

Women workers are currently seeing lower, slower and ultimately less transformative investments in technologies that could improve their work, wages and economic prospects if they valorized their expertise.

By designing for today's workforce, leaders have the opportunity to invest in workforce strategies that improve job quality, productivity and well-being over the long term, rather than intensifying workloads, reinforcing hierarchies and creating decent-work deficits, as the International Labour Organization warns.

Much of the business case surrounding AI is predicated on the prospect of improved productivity and workforce transformation.

Meaningful findings about AI productivity are far from definitive. Early evidence suggests productivity gains in specific tasks, including coding, mid-level professional writing and customer support services, segments where outputs are already digitally mediated and relatively easy to measure.

However, these gains are not universal and are context-dependent. What’s more, the evidence base is still stronger at the task level than at the level of broad economic transformation.

Whether these expectations are outsized or not remains to be seen. Regardless, investment decisions would signal optimism about an AI boom and simultaneously point to a growing tension between economic value and expected financial return.

To date, the AI boom is visible in capital flows and market valuations: investment and market attention remain concentrated on technology providers, AI infrastructure and its digital ecosystems.

This concentration highlights an important feature of the AI economy. While AI may make knowledge and some digital tools more accessible, every AI application depends on expensive infrastructure, including data centres, cloud services, semiconductors and computing power.

These assets require large and sustained investments, making it difficult for individuals and smaller firms to own and engage with them.

As a result, financial gains from AI are flowing to the companies and investors that own the infrastructure, platforms and business models that make AI possible. Groups not represented in this asset-holding class are likely to miss out on the wealth emerging from an expanding AI economy, and gender gaps in wages, wealth and economic power risk widening.

Women are often positioned as users of AI but are scarcely represented among asset owners across the AI value chain. At the very bottom, women are estimated to embody over half of data workers, professionals who clean data, train models and verify outputs.

Yet as value concentrates further up, women’s representation falls significantly. In the US and economies with similar makeups, they are outnumbered two-to-one in high-value segments such as AI talent.

The pattern becomes even more complex when viewed through the lens of capital. Women are increasingly participating in the AI boom, just not taking the same share of winnings.

Pitchbook’s US venture capitalist (VC) female founders dashboard reveals that in 2025, mixed companies with female founders doubled the amount of money invested in them in 2023 to reach nearly $74 billion and about 18.4% of VC capital.

This whopping uptick was AI-driven, as two-thirds of VC money raised went to AI companies. In contrast, exclusively female-founded companies continued to capture just 2% of total funding, while reporting lower value deals and lower deal counts.

As a result, ownership in the AI economy remains highly concentrated among a relatively small group of majority-male-led and male-governed firms, influencing how the technology is designed, deployed and priced.

As AI moves from venture capital to public markets, industry and policy leaders have an opportunity to advance gender parity in value creation, from innovators and founders to investment committees and cap tables.

Women are not entering the AI era with lower capability. In this AI economy, gaps in ownership and investment become gaps in power.

Women’s hesitation toward AI adoption, therefore, may not simply reflect a lack of confidence or interest but rather a reticence to engage with a closed-tiered system offering low, limited rewards.

Industry and policy leaders can shift this equation by ensuring that future investments flow into revaluing occupations predominantly held by women and into increasing the number of women architects and co-owners of the system.