UN urges urgent global action to govern AI risks

Adolescent girls learn computer skills at a primary school in India.

Artificial intelligence (AI) is advancing at a pace faster than governments can regulate, prompting growing concern among experts over the lack of effective global safeguards.

A preliminary report released on Wednesday by the United Nations Independent International Scientific Panel on Artificial Intelligence warns that while the window for establishing effective global AI governance remains open, it may not remain so for long.

The report highlights AI as one of the most transformative technologies in human history, capable of reshaping healthcare, education, scientific research, agriculture and global economies.

Just a few years ago, AI systems were largely limited to answering questions or generating text. Today, they can write software code, analyse massive datasets, create realistic images and videos, accelerate scientific discovery and increasingly perform complex tasks with minimal human supervision.

Experts say the rapid development of AI presents both extraordinary opportunities and serious risks.

Used responsibly, AI could help accelerate progress toward the Sustainable Development Goals by improving healthcare, expanding access to education, strengthening scientific research, boosting agricultural productivity and improving accessibility for people with disabilities.

The report cites several major breakthroughs already made possible by AI.

In healthcare, AI has predicted the structures of more than 200 million proteins, accelerating drug discovery, vaccine development and antibiotic resistance research. Doctors are also using AI tools to detect diseases such as breast cancer earlier, while healthcare workers in developing countries are using AI systems in local languages to improve patient care.

AI is also helping improve food security through early warning systems that identify food shortages before they become crises. In education and accessibility, AI is expanding opportunities for personalised learning and support services.

However, the same technology is also creating major challenges.

The UN panel warns that AI could deepen inequality, spread misinformation, threaten human rights, disrupt labour markets and concentrate power in the hands of a small number of governments and companies.

Among the key risks identified are the spread of deepfakes and sexually explicit AI-generated content, cybercrime, online fraud and disinformation that can undermine public trust and democratic institutions.

The report also raises concerns about mental health, warning that some AI systems may reinforce harmful beliefs or behaviours and contribute to psychological crises.

Experts further caution that increasingly autonomous AI systems may become harder to monitor and control without stronger safeguards.

Environmental concerns are also growing. The energy-intensive data centres powering advanced AI systems contribute significantly to greenhouse gas emissions, adding to global climate pressures.

The report notes that AI development remains highly concentrated among a small number of countries.

The United States controls around three-quarters of the computing power behind the world’s leading AI supercomputers, while China accounts for about 15 percent. Together, the two countries hold roughly 90 percent of global AI computing power.

Most advanced AI models are also being developed by companies based in these two countries.

Meanwhile, many developing nations lack the computing infrastructure, technical expertise, investment and local-language resources needed to fully benefit from AI.

The panel warns that without intervention, AI could widen global inequalities rather than reduce them.

According to the report, existing governance systems are not designed to keep pace with such rapidly evolving technology.

Although more than 40 AI governance frameworks and ethical guidelines already exist worldwide, they remain fragmented, inconsistent and often lack meaningful oversight.

The report calls for stronger independent evaluation, greater international cooperation and common global standards to ensure AI systems remain safe, transparent and accountable.

It also stresses the need for countries to invest in digital infrastructure, education and institutional capacity so they can govern and deploy AI effectively.

The United Nations is supporting efforts to build a new international framework for AI governance.

In 2025, the UN General Assembly established the Independent International Scientific Panel on Artificial Intelligence, comprising 40 experts from different regions of the world serving in their personal capacities.

The panel’s role is scientific rather than regulatory. It regularly assesses evidence on AI’s opportunities, risks and impacts and produces independent reports to support policymaking.

Its findings will contribute to the UN Global Dialogue on AI Governance, scheduled to begin in Geneva on July 6, 2026, where member states will discuss international approaches to managing AI.

The report concludes that AI itself is neither inherently good nor bad.

Its long-term impact will depend on the decisions governments, companies and societies make today.

Whether AI reduces inequality or deepens it, and whether it strengthens or weakens democracy and human rights, will largely depend on how quickly the world can build governance systems capable of keeping pace with innovation.