Europe’s AI talent is slipping away. Building a sustainable AI education ecosystem depends on better career prospects and stronger public-private efforts.
The European Commission’s Directorate-General for Research and Innovation (DG RTD) recently released a report describing the shortages of Artificial Intelligence (AI) education and talent in science and research in Europe, exploring the roots and causes, as well as the potential political and institutional remedies. This report belongs to a series of reports within the Mutual Learning Exercise (MLE) on national policies for AI in science, sharing best practices from numerous Member States (MS). Over the past years, Europe has experienced a steady brain drain of AI professionals, with many relocating to the US, followed by destinations like China, Australia, Canada, and the UK. European graduates in AI-related fields often outflow Europe as early as to pursue their postgraduate studies abroad. The ability to retain top AI talent is now a strategic imperative for the European Union (EU), which aims to harness the potential of AI in scientific research, recently releasing its AI Continent Plan (see SwissCore article). AI-driven science relies on interdisciplinary expertise, including computer science, statistics, mathematics and domain-specific knowledge. Without a strong talent base, the teaching, development and application of AI in science is not possible. Talent retention is also crucial for national and European competitiveness, scientific progress and autonomy. The report lists four main barriers to developing AI talent for science identified by Member States participating in the MLE.
The retention of AI talent is one of the main challenges for Europe. This involves not only retaining talent within its borders but also within the research sector. Namely, the academic and public sectors face multiple challenges, particularly when compared to the more attractive conditions offered by the industry. Higher salaries, access to advanced resources (e.g. proprietary datasets not accessible through academic channels) and broader career prospects and improvements make tech companies, and increasingly the financial sector, more appealing to AI and information systems graduates. Academia also often lacks a long-term strategy to retain researchers, with careers heavily reliant on evaluation systems focused on publication metrics and difficult reintegration after time spent in the private sector. However, opportunities exist to reverse this trend: strengthening collaboration between academia and industry, developing more flexible career pathways and establishing strong public-private partnerships can help bridge the gap. Policy interventions are also needed, such as simplified visa processes to promote international mobility and increased funding for AI research.
The report states that AI talent must be fostered not only at the tertiary level but also from an early age. AI education must be systematically integrated at all levels of the system, from primary schools to universities. The study reports a quality of AI education varying significantly across countries, underlining the need for coordinated national strategies and agile education systems that can adapt quickly to technological changes. The introduction of AI concepts at an early stage of education also helps foster interest and develop foundational knowledge. Interdisciplinary curricula integrating AI education are needed at all levels of education and are crucial in preparing students. Particularly at the tertiary level, students from diverse academic backgrounds would benefit from being exposed to AI education. AI research often faces significant structural barriers within academic institutions, where traditional silos can hinder collaboration. Academic institutions must encourage cross-disciplinary and cross-sector partnerships, while national funding organisations have a crucial function in supporting collaborative research. Public-private partnerships play an important role in offering resources and practical expertise that can enhance educational content and ensure that the latter is aligned with labour market needs.
The report suggests that upskilling the existing workforce is also essential to harness the potential of AI. For the talent pool to stay competitive, training is crucial. Currently, many researchers lack formal training in applying and working with AI models in their research. Building AI literacy across all disciplines would further strengthen the research ecosystem and enable researchers from all backgrounds to engage with AI. The report advises universities to integrate AI and machine learning courses into STEM curricula, making those courses essential components of all scientific domains, including in PhD programmes. The creation of tailored programmes for underrepresented communities and ensuring gender balance in AI education are also crucial for the sustainability of the education model.
To remain competitive in research, retaining talent in AI science is a strategic priority. Achieving this requires an integrated approach. Education and training continue to remain at the core of AI talent development and policymakers must take a proactive role by embedding AI as a priority of national research strategies. Higher education institutions have a role to play in fostering AI upskilling and coordinated efforts across sectors are needed to build an effective AI workforce for scientific purposes. While challenges persist, academia continues to offer valuable aspects that many researchers appreciate, with intellectual freedom and academic independence on top of the list. These are often less prevalent in industry settings. By recognising these strengths through better incentives and support, academic institutions can make research careers in AI more attractive and reduce AI talent migration to the private sector or to other countries.