AI as both a driver of skills demand, and a solution to skill supply While demand for digital skills is likely to remain strong, AI may eventually reduce demand by automating tasks. Cognizant research suggests up to 90% of jobs could be affected by AI by 2032, with some roles, like computer developers, facing significant disruption.14 However, predictions about the future are mixed. For example, Google’s 2024 DORA report found only slight improvements in code quality and review speed from AI tools, but also noted decreased delivery stability and throughput.15 One study even showed that experienced open-source developers became 19% slower when using AI tools, challenging assumptions about efficiency gains.16 At the same time, AI offers new opportunities that could speed up skill development (Box 4). For instance, AI-powered personalized learning platforms can adapt content and pace to individual learners, allowing beginners to quickly grasp core concepts in areas like AI literacy, programming or data science. Interactive virtual tutors and coding assistants, such as AI-driven code editors or chatbots, provide instant feedback and guidance, helping learners solve problems more efficiently. Additionally, AI can curate recommended learning pathways, highlight relevant resources and even generate tailored practice exercises, streamlining the acquisition of both technical and human-centric skills. The rapid advancement of genAI has prompted debate about its impact on work, reshaping not only how tasks are performed but also the future relevance of skills, particularly digital skills. Analysis by Indeed highlights both the growing capability of AI tools to transform digital skills and the rising importance of human proficiency with digital tools and technologies in an AI-driven economy. Figure 11 presents the potential for genAI to transform work skills, drawing on Indeed Hiring Lab’s GenAI Skill Transformation Index. The index scores skills across cognitive abilities and physical requirements, and measures how AI could change the way skills are used or work is done. Using the World Economic Forum’s Global Skills Taxonomy, the analysis classifies nearly 2,900 granular work skills into four categories of transformation potential under genAI: minimal transformation, assisted transformation, hybrid transformation and full transformation.17 AI, data and digital skills are the most exposed to transformation, while human-centric skills are expected to see relatively minimal impact. Taken as an average, 68% of digital skills will see either hybrid or full transformation, compared to 35% across for all other, more human-centric, skills. At the greatest extremes this gulf is vast: AI and big data skills, for example are over 30 times more likely to see full or hybrid transformation compared to empathy and active listening. AI and programming skills show greater capacity for transformation, as genAI can already handle many routine tasks and even perform independently in areas such as text or image classification, sentiment analysis, data preprocessing and prompt engineering. Still, this does not mean human workers will be displaced. The pace and extent of change will depend on model capabilities, organizational adoption, regulatory frameworks and the specific context in which tasks are performed. By comparison, technology literacy – the ability to apply digital tools to context-specific business or social challenges – is far less affected. Many of these tasks rely on human judgement, creativity and adaptation. This divergence underscores the need for a dual investment in skills: advanced AI and data capabilities to manage and oversee digital systems, and a broad-based digital fluency that enable all workers to adapt, apply and reshape technology to address real-world challenges. How genAI is transforming digital skills BOX 4 New Economy Skills: Building AI, Data and Digital Capabilities for Growth 19

New Economy Skills 2025