Demand for AI talent continues to outpace supply. Nearly half of enterprise leaders cite skills gaps as a major barrier to AI adoption, according to a 2025 McKinsey survey. Meanwhile, AI’s rapid pace of development is shrinking the half-life of many skills. A 2025 World Economic Forum report found that employers expect around 40% of workers’ core skills to change between 2025 and 2030, with AI and technological literacy becoming increasingly important. Globally, enterprises, educational institutions, and governments all face a tough battle to develop the current and next generation of AI talent.
Our research indicates enterprises face two key types of AI skills shortages: technical and foundational.
Go deeper with GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
Technical AI skills shortages
Technical AI specialists, including AI engineers, AI architects, AI research scientists, and AI governance leads, remain scarce and highly sought after. GlobalData’s Job Analytics database shows active AI-related job postings increased 85% between January 2024 and January 2026.
AI hyperscalers and well-funded AI start-ups dominate the competition for top technical AI talent by offering astronomical compensation packages and compelling work at the AI frontier. Retention is proving difficult due to aggressive competition, which is pushing up compensation packages even further. Pure-play AI vendors such as OpenAI and Anthropic face greater pressure to secure top technical AI talent, as their survival relies solely on AI. According to The Wall Street Journal, an unprecedented 46% of OpenAI’s projected 2025 revenue will go to stock-based compensation, with an average of $1.5m in stock-based compensation across its roughly 4,000 employees. This illustrates how aggressively some AI firms are using equity to secure top technical AI talent. However, such large equity awards are inflating operating losses and diluting existing investors’ ownership stakes.
Foundational AI skills shortages
Foundational AI skills shortages are arguably more rife, covering the core and role-specific AI knowledge that employees require to use AI effectively and safely. While access to generative AI tools is widespread, enterprise AI training programmes have generally been too generic, infrequent, or absent altogether, preventing AI from meaningfully changing how work gets done. Agentic AI’s rise has added additional AI training needs for broader workforces, such as understanding how to provide oversight over AI agents.
Workers recognise the gap. Only 29% of workers globally say their workplace invests enough in AI training, according to a 2025 Salesforce and Morning Consult survey. Some proactive workers are pursuing AI upskilling independently, through platforms like LinkedIn Learning, to compensate for weak internal programmes.
How enterprises can fill AI skills shortages
Enterprises require an AI talent strategy to ensure they have the right skills to deploy AI competitively and responsibly.
A good starting point is developing an AI skills taxonomy that maps the technical and foundational AI skills needed organisation-wide to deliver on the company’s broader AI strategy. Enterprises can then use this taxonomy as a benchmark to assess current AI skills, pinpoint gaps, and implement internal and external measures to fill them.
In terms of internal measures, enterprises should invest in strategic AI upskilling and reskilling, with ongoing core and role-based training for all workers with AI exposure. They should also build AI champion networks that act as a conduit between central AI teams and business units, helping model AI adoption, coaching peers on AI use, and surfacing AI adoption barriers that might otherwise not reach senior leadership. Establishing structured, ongoing engagement between the central AI team and the AI champion community will prevent the network from becoming a symbolic initiative.
In terms of external measures, enterprises might look to recruit an experienced and esteemed AI professional to steer their AI strategy and act as a magnet for other AI specialists. They could also lean on AI consultants and contractors to fill critical AI skills gaps and maintain momentum on AI initiatives while internal capabilities are being developed. They should also strengthen early talent pipelines by partnering with universities through internships, graduate programmes, and collaborative research to access the next generation of AI talent. Some enterprises have already cut entry and graduate roles due to AI, but this is a short-sighted approach that risks eventual mid-level talent shortages and weakens skill foundations.
Enterprises must also ensure workforce planning is more data-driven. AI-powered workforce planning tools can help monitor and forecast AI skills needs. For example, IT services provider Wipro used Edligo’s AI-powered talent analytics platform to create a unified skills inventory of employees across 16 countries.
Alongside hard AI skills, soft skills that AI cannot easily replicate, such as leadership, empathy, and collaboration, remain paramount for employers. Soft skills will be an important competitive differentiator for enterprises as foundational AI skills proliferate.
