Skills in the AI age

To make the most of benefits from AI technologies while mitigating the risks, governments must put into action agile policies that work in tandem (OECD, 2024[36]).

As AI transforms industries and job requirements, investing in the quality of education and training systems is critical to ensure learners acquire both foundational and advanced digital competencies. Many governments allocate funding programmes that focus on in-demand skills and technologies. These programmes support vocational training, apprenticeships and on-the-job training opportunities and provide the relevant training to teachers. For example, Latvia’s “Future Skills for Future Society” initiative advances the goals of Education 4.0 by fostering digital competence, innovation and lifelong learning. This ensures that education and training systems remain agile and aligned with the evolving demands of an AI-driven economy. Likewise, “Education 4.0. CARNET Strategy” in Croatia emphasises integration of digital technologies, AI literacy and teacher upskilling to create an inclusive, future-ready education system that bridges learning with labour market needs.

In 2023, Spain launched a Microcredentials Plan funded by the European Union (EU) Recovery and Resilience Facility to support adult reskilling and upskilling for learners aged 25-64. Public funding covers 70% of course costs, while learners or employers pay the remaining 30%, encouraging training relevant to the labour market. While not limited to digital skills, this model has led to a strong presence of AI-related microcredentials in the 2025-2026 offerings.

High-quality curricula should integrate AI literacy (see below), critical thinking, data analysis and problem solving across all levels of education. Finland’s National AI Strategy promotes AI education from primary schools to universities, ensuring students understand AI’s ethical and practical implications. Similarly, the European Commission’s Digital Education Action Plan (2021-2027) calls for upgrading teacher training and digital infrastructure to maintain educational excellence in the face of technological disruption. Launched in 2025, France’s Osez l’IA plan aims to build digital and AI capabilities across firms via an Académie de l’IA platform. It offers training and tutorials for executives, employees and entrepreneurs, helping potential founders gain the skills needed to start and scale AI-based ventures.

Many countries are also investing in real-time labour market data and sophisticated skills forecasting models to better predict future workforce needs. Australia is working with data analytics companies to analyse real-time trends in job advertisements. This initiative aims to identify skills that have grown in demand in recent years by matching information from job advertisements to detailed occupational profiles within the Australian Skills Classification. Such analyses provide granular insights into evolving skill requirements across different sectors.

Training programmes to develop high-skilled ICT professionals sometimes entail the creation of new bachelor, master or doctoral programmes. Poland, as part of the “Academy for Innovative Applications of Digital Technology (AI Tech)” initiative, has co‑ordinated a collaboration between leading companies in the technology sector and Polish universities to jointly create AI-related master programmes. Five universities have launched new second cycle programmes and specialisations in AI, cybersecurity and machine learning in 2021, with funding from the EU (OECD, 2024[37]).

Some countries have developed short, non-formal training programmes aimed at upskilling or reskilling employed or unemployed adults to cope with changes brought about by the AI transition. In 2020, the United Kingdom Department for Education launched “Skills Bootcamps”, which offer short sector-specific skills training to both employed and unemployed adults in different areas, including digital and AI. For instance, a Skills Bootcamp on Data/AI helps learners develop the technical and employability skills to secure an entry role as a data technician or analyst. Course curricula are created and delivered in collaboration with employers. The courses last up to 16 weeks and are free for the participant. Upon successful completion of a Skills Bootcamp, participants are offered a job interview with a potential employer, helping workers make the transition from training to work (OECD, 2024[37]).

To support workers most at risk of displacement due to AI and automation, governments also need to expand lifelong learning and upskilling programmes. Despite growing consensus on the importance of lifelong learning for economic resilience and individual opportunity, participation in adult education is stagnating or even declining in many countries (OECD, 2025[38]).

Yet, many countries continue to introduce or expand innovative programmes. Singapore’s “SkillsFuture initiative” for example provides financial credits to individuals for continuous training and offers AI-specific learning pathways co-designed with employers. Likewise, Denmark has created tripartite training councils that bring together employers, unions and government to shape reskilling efforts in response to technological change, including in manufacturing and logistics.

In Finland, the widely recognised “Elements of AI” online course, developed in partnership with the University of Helsinki, is freely available to all citizens. It has been translated into multiple languages, fostering digital inclusion and lifelong learning at scale. Korea’s “Artificial Intelligence & Digital (AID) 30+” project strengthens lifelong learning opportunities in digital and AI-related fields for adults in the prime working-age population (Box 1).

Expanding flexible learning options (e.g. online, part-time) is key to broadening accessibility and inclusivity of AI training, increasing the attractiveness of training and encouraging wider participation among underrepresented groups (OECD, 2023[40]). Modular training programmes can enable adults to learn in their own time and shape their own learning path. Switzerland is a front-runner for modular training opportunities where such programmes are available throughout the adult learning sector. Adults can take part in individual modules or combine different modules to form a full training programme (OECD, 2019[41]).

The validation or recognition of prior learning can also benefit individuals, employers and the economy as a whole. It makes all knowledge visible, including the diverse learning that takes place outside of formal education and training. It also allows for a better match between jobs and workers’ skills (OECD, 2023[42]).

Encouraging employer-led training is another key strategy to encourage lifelong learning. The rationale for supporting employer-based training is twofold. First, while firms benefit from a more skilled workforce, they may underinvest due to concerns about employee turnover or uncertain returns to training. Second, employer-supported training remains the dominant form of adult learning, and financial incentives can help shape firm behaviour to promote upskilling (OECD, 2025[38]). Public financial incentives (such as training leave, individual learning, accounts and targeted subsidies to employers) are instruments that may reduce barriers to participation, particularly for those on low incomes or doing non-standard work (OECD, 2025[38]). It is equally important that employers help shape public-led training.

With respect to engaging employers in training, Germany provides a valuable model. Through its “Qualification Opportunities Act” (Qualifizierungschancengesetz), Germany subsidises wages and provides support for training costs to employers investing in the upskilling of workers at risk of redundancy due to digital transformation. The funding system for promoting continuing vocational education and training (CVET) – regulated in Book III of the Social Code – provides the Federal Employment Agency (Bundesagentur für Arbeit) with tools for employees and employers. Support options include covering continuing education costs (e.g. for training courses, accommodation and childcare, and travelling expenses) or providing wage subsidies for employers. The promotion is open to sectors or topics beyond digital technologies (Box 2).

Austria and the Netherlands have adopted similar co-financing initiatives to incentivise in-company training in AI and digital technologies. Likewise, France’s “Individual Learning Account” (Compte Personnel de Formation, CPF), provides individuals with personal learning accounts funded by the state. This allows workers (including those in non-standard or low-income employment) to accumulate training credits and access subsidised upskilling opportunities throughout their careers.

AI literacy can be defined as a set of competences that enables individuals to critically evaluate AI technologies; communicate and collaborate effectively with AI; and use AI as a tool online, at home and in the workplace (Long and Magerko, 2020[46]).

Most workers will only need literacy to help them use and interact with AI systems effectively. Thus, offering training programmes that promote literacy are equally important as those focused on advanced AI skills. AI literacy training aims to equip workers with a general understanding of AI, enabling them to communicate and collaborate effectively with AI technologies. It can involve training on AI ethics, fostering awareness of potential risks and harms, and equipping individuals to critically evaluate AI technologies as they use them.

While many countries have introduced training incentives with the recent rise of AI adoption, training often lacks a direct focus on AI and AI literacy. A survey of 21 OECD Member countries shows 14 have invested in AI-specific publicly funded training programmes. Of these, nine focus on developing AI professionals, while seven aim to build AI literacy among the general public (OECD, 2024[37]).

To promote international comparability in this area, the OECD and the European Commission are developing an AI literacy framework that outlines the knowledge, skills and attitudes that will adequately prepare students in primary and secondary education. The initiative will also provide the foundation for the first assessment of AI literacy in the 2029 Programme for International Student Assessment (PISA). Furthermore, it will support EU goals to promote quality and inclusive digital education and skills (Schleicher, 2025[47]).

Countries rely heavily on financial incentives to promote development of AI skills, underusing other policy tools that could support a more inclusive transition. Alternative tools such as career guidance initiatives, public-private collaborations and train-the-trainer programmes reach a wider and more diverse audience than financial incentives to develop AI skills (OECD, 2024[37]). Evidence from selected countries also suggests training supply may be insufficient to meet demand, especially for AI literacy.

Adapting the content of training supply to deliver AI skill needs calls for linking financial support measures (such as subsidies and grants) more directly to general AI literacy and AI competencies, with particular attention to those most vulnerable to transition risks. Non-financial incentives should also be strengthened through, for example, support for career guidance, public-private collaborations and train-the-trainer programmes. Collaboration between governments, educational institutions, and industry will help co-design curricula and training courses that are well aligned with current and future workplace needs. Expanding flexible (e.g. online, part-time) and modular learning options will help make training more attractive, and broaden the accessibility and inclusivity of AI training.

To build AI competencies across the workforce, early and continuous investment in education and skills development is critical. This begins with integrating digital and AI literacy into general education from a young age and extends through vocational education and training (VET) systems that are closely aligned with evolving labour market demands. For instance, Estonia’s “AI Leap” initiative incorporates AI-focused curricula into both general secondary and vocational education, preparing students with foundational skills in AI tools, ethics and critical thinking (Box 3).

Technological advancements, such as AI and automation, lead to shifts in the labour market and may displace workers. Targeted support for these workers, including retraining programmes and re-employment assistance, can help mitigate long-term economic losses. This can also help prevent vulnerable groups and regions from being left behind amid technological changes (OECD, 2024[17]).

Belgium supports displaced steelworkers in Liège through tailored upskilling and reskilling. With EUR 2.65 million in funding (including support from the European Globalisation Adjustment Fund), the programme provided displaced steelworkers with a range of targeted active labour market measures. These range from short skill upgrades to full retraining pathways and a transfer of experience that enabled veteran workers to become technical instructors.

Finland’s “Bridge Programme” was designed to support workers displaced by Nokia’s restructuring. It combined financial incentives, mentorship and access to entrepreneurial networks to encourage early career transitions and self-employment for affected workers.

Unemployment insurance systems in OECD Member countries show emerging patterns. These include the integration of income support with active labour market services, variations in benefit generosity and duration, and the role of insurance programmes in smoothing income losses during labour market transitions. OECD (2025[49]) underscores the importance of designing unemployment insurance that provides financial stability and also interfaces effectively with re-employment services.

Despite policy efforts to widen access to adult learning, participation remains highly unequal. OECD (2025[38]) shows that older workers, migrants and those with lower formal qualifications are significantly less likely to participate in training than other groups. These disparities reflect both structural barriers such as cost, time constraints and accessibility, and lower perceived benefits of training among these groups. In addition, individuals who have had negative experiences of formal education may lack confidence in their ability to learn, while others may struggle to find training opportunities that meet their specific needs.

Addressing these inequalities requires more than targeted initiatives. In many cases, adult learning systems as a whole fail to meet the needs of disadvantaged groups. If systems remain overly complex, inflexible or employer-dependent, even well-designed outreach and targeted financial incentives will have limited impact. Ensuring access requires rethinking adult education systems, including simplifying learning pathways; embedding more flexible, learner-centred approaches; and enabling adults to train for skills in demand (OECD, 2025[38]).

In Italy, “Provincial Centres for Adult Education (CPIA)” support low-skilled adults and migrants whose qualifications are not recognised. Their programmes include literacy and Italian language courses, lower secondary education and pathways leading to upper secondary diplomas. Learners can have their competences assessed and validated, reducing the number of hours they need to complete their studies.

In Canada, the “AI4Good Lab” programme offers AI training and mentorship for women and disadvantaged groups, addressing both the skills and representation gaps in emerging technology roles (Box 4).

Similarly, Australia’s “Women in STEM Ambassador Programme” works with schools, universities and industry to dismantle barriers to equity through inclusive hiring practices, mentoring networks and awareness campaigns.