Preparing the workforce for the challenges of AI

The OECD experience

In April 2025, the OECD released a report¹ on training and adoption of AI technologies. About one-third of jobs in OECD countries involve AI, and only 1% of jobs require highly specialized AI competencies (like skills in developing, customizing AI systems: machine learning, building and training models, working² with big data, using frameworks (like³ TensorFlow and PyTorch )).

The professions most affected by AI are those where more than 70% of employees have higher education - managers, IT specialists, scientists, accountants and translators. More⁴ than 90% of them are of working age. However, there is no evidence of mass displacement of such workers as a result of AI⁵ implementation. In general, for most workers, basic knowledge and skills in interacting with AI (“AI literacy”) are sufficient: the ability to use AI tools in everyday tasks, understand their algorithms and assess the risks associated with their use.

Analysis of educational programs has shown that in Australia, Germany, Singapore and the US, only 0.3-5.5% of courses include modules on AI. The main focus is on training specialists in AI development, while programs to develop general AI literacy are less common.

The following trends in OECD countries can be identified:

– Building basic digital and AI literacy among a wide audience. For example, Austria is implementing the Digital Everywhere project: in 2024, 3,500 master classes on digital skills (including AI and cybersecurity) were held.

– Promoting AI and digital solutions for SMEs.

– Retraining workers at risk of automation. Singapore implements two-day courses on automation, cyber risk and analytics for low-skilled workers (in the food, textile and manufacturing industries), who are predominantly men without tertiary education and migrants.

– Training highly qualified specialists in AI and related fields. The UK implements Skills Bootcamps - 16-week courses on AI and digital technologies in cooperation with employers.

The OECD recommends developing AI skills through the following measures:

−Financial support: Subsidies, tax credits, training vouchers and grants for employers, especially for training workers at risk of automation.

−Non-financial measures: Career counseling, development of teacher training programs, creation of partnerships between universities and businesses.

−Lowering access thresholds: Simplifying entry requirements for AI courses.

−Developing general AI literacy initiatives through specialized short intensive courses.

−Integrating AI learning into HR strategies.

Russia’s experience

In April 2025, a Rosbalt analysis showed that at least 10 mln Russians (cashiers, machine operators and middle-aged white-collar workers)⁶ are at risk of unemployment. For example, FixPrice launched 5,800 self-service terminals, ensuring the processing of about a third of all transactions. Due to similar automation in other retail chains (Magnit and X5) around a million⁷ cashiers may be out of work.

Russia is developing free or partially subsidized courses on digitalization. It is planned to train 600 thousand people by 2030, the Ministry of Digital Development, Communications and Mass Media is selecting organizations and programs to receive⁸ subsidies.

From the monitoring issue No. 4 (16), April 2025. Download the full issue (PDF) · issue page at the Gaidar Institute