The World Economic Forum (WEF) and LinkedIn published a new white paper, Gender Parity in the Intelligent Age (March 2025). It draws on insights from leaders in industry, policy and multilateral organisations.
Human Level’s Take:
- The World Economic Forum and LinkedIn make the business case for inclusive AI practices clear: “Companies that fail to integrate gender parity into AI strategy will miss out on half of the available talent, reducing their capacity for innovation and long-term competitiveness.”
- A diverse talent pool is essential for innovation. This means that economies advancing in AI without diversity may face setbacks and inequality, while those attracting diverse talent will gain a competitive advantage.
- In addition, AI is impacting jobs and career paths differently for men and women, with women more likely to hold roles disrupted by AI. Although women’s participation in tech is growing, retention remains a challenge, and men still dominate STEM leadership roles. De-biasing AI hiring tools will be an essential step to reach gender parity in these fields.
- There are a few key pathways for companies to advance gender equity as they pursue a tech transition. This includes promoting parity in skilling and reskilling, eliminating workplace bias; fostering an inclusive culture; and enhancing diverse representation in data to improve learning models and outcomes.
Some key takeaways:
- Gender diversity offers a competitive edge in tech transformation: The paper finds that gender gaps are changing in the context of emerging technology and that gender diversity can be a boon to the technological transition. Technology is playing a significant role in society, with artificial intelligence (AI) technologies being developed to tackle various economic challenges, such as job creation, productivity growth and GDP enhancement. Economies that tap into the widest pool of talent during this transition will be in the best position to achieve a resilient, innovative, and successful shift into the intelligent era. A diverse talent pool is crucial for innovation, but women often drop out of the talent pipeline at various stages. Uneven innovation ecosystems further concentrate women innovators in only a few economies. The authors suggest that economies advancing in AI with limited diversity risk facing economic setbacks and inequality. As AI progresses, economies that attract diverse talent will have a competitive edge.
- AI affects men and women differently in the world of work: AI affects jobs and career paths differently for men and women. Women are more likely to hold roles disrupted by AI (for example, those using skillsets like communication, decision-making and relational tasks) and are less likely to experience job augmentation. Despite this, more women are acquiring AI skills, and their participation in tech has grown to nearly one-third. However, retention remains a challenge, and men still dominate in STEM leadership roles. Research suggests that AI's evolving nature offers an opportunity to address gender disparities. From 2018 to 2025, female AI talent on LinkedIn has significantly increased, narrowing the gender gap in 74 out of 75 economies, with underreporting possibly indicating a larger pool of female talent. When it comes to AI “augmentation”— such as using AI for hiring — there can be risks of bias that undermine fair and inclusive hiring. The paper reports that 99% of Fortune 500 countries are using automation in their hiring practices. This means that addressing gender bias in AI is a critical step to ensure that women can also benefit from AI-driven career opportunities.
- How business leaders can take action: The paper identifies a few key ways in which business leaders can take action to ensure their companies benefit from gender diversity and parity in AI. This includes addressing gender disparities throughout the design, development and adoption of AI technology; ensuring parity in skilling and reskilling for the tech transition; taking up practices that remove bias in workplaces and shifting the culture towards inclusion; increasing female participation in the tech transition accounting for changing populations; ensuring parity in resourcing for the transition; and increasing representation of diverse groups in data, in order to improve learning models and their outcomes.
