UN warns of growing inequality due to AI - ForkLog

# UN Warns of Rising Inequality Due to AI

The rapid development of artificial intelligence could exacerbate global inequality. This warning is contained in a preliminary report from the UN Independent International Scientific Panel on AI.

The document was prepared by 40 independent scientists and experts. In their view, access to such tools alone does not guarantee equal benefits. Countries that depend on foreign models, cloud infrastructure, and data flows may gain access to the technology but lose practical control over its standards, safeguards, and local adaptation.

According to the authors, the main problem is the concentration of computing infrastructure and the development of advanced models. They report that the United States controls about three-quarters of the computing power of the world's leading AI supercomputers, while China accounts for approximately 15%. Together, these two countries account for about 90% of the computing used to train the most powerful systems.

The UN warned that the concentration of AI capabilities among a small number of companies and states could strengthen authoritarian control and weaken democratic accountability. As the authors noted, lagging countries need to develop their own computing and data infrastructure, specialists, and local developers.

"In the long term, the gap between rapidly growing functional capabilities and the availability of effective risk management methods could lead to catastrophic results," the report states.

A separate section is devoted to language inequality. According to the researchers' conclusions, generative models perform better with English and other widely spoken languages, while others are either barely represented or receive noticeably worse quality.

The report gives an example of machine translation from the Tigrinya language: the system confused smallpox with syphilis, gonorrhea with diabetes, and the phrase about intravenous antibiotics with "intravenous insecticides."

"Such translation errors can be life-threatening," the report's authors emphasized.

A separate factor of inequality is basic internet access. According to the International Telecommunication Union, in 2025, 2.2 billion people remained offline. For these regions, the introduction of AI first hinges on access to the internet, energy, and digital skills.

The preliminary report is not a regulatory document and does not establish binding rules. The panel provides a scientific assessment of AI's capabilities, risks, and consequences, which governments can use in policy development. Among the directions highlighted by the authors:

• development of local AI infrastructure and data centers;
• improving AI literacy in schools and workplaces;
• investment in developers and technical expertise;
• creation of AI safety institutions;
• strategies to combat disinformation;
• testing of AI systems after launch on real users, tasks, and conditions.

The report also pointed out risks of disinformation, fraud, election interference, human rights violations, and pressure on labor markets.

Recall that in June, the Bank for International Settlements concluded that the investment boom around artificial intelligence, which supported the global economy in 2025, is itself becoming a source of macro-financial risks.