A new report from a United Nations agency warns that AI is widening global income inequality and outlines a plan to harness its potential more equitably.
In addition, the Technology and Innovation Report 2025 by the UN Conference on Trade and Development (UNCTAD) predicts that artificial intelligence could affect up to 40% of jobs worldwide through automation and job displacement. These changes could especially impact countries that rely on low-cost labor for competitive advantage.
“History has shown that while technological progress drives economic growth, it does not on its own ensure equitable income distribution or promote inclusive human development,” said Rebeca Grynspan, the Secretary-General for UNCTAD.
Concentration of AI global investments — and who is excluded
The UN report projects that AI is on track to reach a market value of $4.8 trillion by 2033; however, most of this wealth and technological power remains concentrated in a handful of companies and countries.
According to the report, only 100 companies are behind 40% of global investments in AI companies — and most are based in the U.S. and China — resulting in significant geographic consolidation. In contrast, 118 countries, primarily in the Global South, are largely excluded from global AI governance discussions.
Possibly good news about AI’s impact on jobs
The report isn’t entirely pessimistic; it emphasizes that AI won’t simply replace jobs but can also create new opportunities and “empower workers.” UNCTAD says that if countries and governments invest in workforce adaptation through reskilling and upskilling, it could prevent AI from eliminating too many jobs.
How countries can prepare workers for the AI economy
UNCTAD says that developing countries need to invest in three key leverage points to avoid being left behind in the AI economy; those three points are infrastructure, data, and skills.
- Infrastructure: Investing in fast and reliable internet, as well as the computer power needed to run AI models and store training data.
- Data: Training AI models effectively with diverse, high quality data sets.
- Skills: Creating education systems and teaching employees the skills they need to succeed in an AI-driven workforce.
International AI collaboration is more important than ever
In the report, UNCTAD says that greater collaboration is also necessary between countries, instead of only placing the burden on individual governments. The report proposes the creation of a shared global facility that would ensure equitable access to AI tools and computer power for all countries.
To boost accountability and transparency, the report suggests creating a public disclosure framework for AI. A similar disclosure framework also exists for environmental, social, and governance (ESG) standards, which could be used as a model.