Just a few years ago, it could answer questions or generate text. Today, it can write computer code, analyse vast amounts of data, create realistic images and videos, help scientists discover new medicines and increasingly act on its own with little human supervision.
However, while AI’s capabilities are accelerating, experts say that the rules ensuring AI is used safely are struggling to keep pace.
That is the conclusion of the preliminary report by the UN Independent International Scientific Panel on Artificial Intelligence launched on Wednesday.
It warns that the window to establish effective global governance remains open but may not stay that way for long.
AI is contributing to major medical breakthroughs.
Why it matters
AI could become one of humanity’s most transformative technologies.
Used responsibly, it could accelerate progress towards the Sustainable Development Goals by improving healthcare, education, scientific research, agriculture and accessibility for people with disabilities.
But without safeguards, the same technology could deepen inequality, spread misinformation, threaten human rights, disrupt labour markets and place powerful AI systems in the hands of very few governments and companies.
The challenge, according to the report, is finding a way to unlock AI’s enormous benefits while preventing its growing risks.
Extraordinary pace of development
AI capabilities have advanced at an extraordinary pace over the past few years.
Powerful new computing networks, vast amounts of training data and improved AI techniques have produced systems capable of fluent conversation, advanced scientific reasoning, software development and creating highly realistic images, audio and video.
The next wave is already emerging.
Instead of simply responding to prompts, AI “agents” can increasingly plan tasks, use digital tools, write software and complete complex assignments with little or no human oversight.
Researchers say the complexity of tasks these systems can complete has been doubling every few months, according to the report.
AI-powered robotics are increasingly being used in agriculture.
The benefits: What can AI do?
The UN report highlights a growing list of real-world successes.
- Medical breakthroughs: AI has predicted the structures of more than 200 million proteins, accelerated drug discovery, vaccine development and research into antibiotic resistance.
- Better healthcare: Doctors are using AI to detect diseases such as breast cancer earlier, while health workers in developing countries are using AI tools in local languages to improve patient care.
- Food security: AI-powered early warning systems are helping to identify food insecurity before it becomes a crisis.
- Improving lives: AI is supporting scientific research, making technology more accessible for people with disabilities, and expanding opportunities for personalised education and mental health support
The panel stresses that these are not future possibilities: They are already happening.
A data centre in the US state of Wisconsin.
The risks: What worries experts?
The same technology is also creating new dangers.
- Online abuse: AI is fueling the spread of sexual abuse material and sexually explicit deepfakes, with women and children most at risk.
- Disinformation: AI can generate false information that is as convincing as the truth, undermining trust in public debate and democracy.
- Crime: Criminals are using AI to carry out cyberattacks, fraud and social engineering scams.
- Mental health: Some AI systems can reinforce harmful beliefs or behaviours, leading to mental health crises, including suicide.
- Loss of control: As AI becomes more autonomous, experts warn it could become harder to monitor and govern without stronger safeguards.
- Environmental impact: The energy-hungry data centres which power AI are contributing to greenhouse gas emissions which leads to global warming.
Who benefits and who could be left behind?
The AI revolution is far from equal.
While it is used around the world, access remains heavily concentrated in developed countries.
The report notes that the United States possesses around three-quarters of the computing power behind the world’s leading AI supercomputers, while China accounts for around 15 percent, giving the two countries roughly 90 percent of that computing power combined.
Most advanced AI models are also being developed by companies based in those two countries.
Many developing countries lack the computing infrastructure, technical expertise, data, investment and local-language resources needed to fully benefit from AI.
As a result, they often depend on technologies they cannot build, inspect, audit or adapt to their own societies.
The panel warns that unless these gaps are addressed, AI could reinforce existing global inequalities rather than reduce them.
A partially sighted student in Uganda uses an assistive device for reading and recording lessons.
Why does AI need regulation?
According to the UN panel, today’s governance systems were not designed for technology evolving this quickly.
Governments face what experts describe as an “evidence dilemma”: Policymakers need reliable scientific data before introducing regulations, but by the time enough exists, the technology may already have moved on.
Although more than 40 AI governance frameworks and ethical guidelines already exist in different parts of the world, they remain fragmented, inconsistent and are rarely tested to see whether they actually work.
Many safety assessments are also conducted by the companies developing the technology themselves.
The report finds that stronger independent evaluation, international cooperation and common standards are needed to ensure AI systems remain safe, transparent and accountable.
At the same time, countries need investment in digital infrastructure, education, technical expertise and institutions so they can govern and deploy AI on their own terms.
What is the United Nations doing?
The United Nations is supporting a new international architecture to help countries make informed decisions about AI.
In 2025, the UN General Assembly established the Independent International Scientific Panel on Artificial Intelligence, made up of 40 experts from every region of the world serving in their personal capacity.
The panel’s role is scientific rather than regulatory. It assesses, on a regular basis, the latest evidence on AI’s opportunities, risks and impacts and produces independent reports that governments can use when developing policy.
The panel’s work will feed into the UN Global Dialogue on AI Governance which begins in Geneva on 6 July 2026, where Member States will discuss international approaches to managing the technology.
The bottom line
The scientific panel is clear: AI is neither inherently good nor bad.
Its impact will depend on the choices governments, companies and societies make today.
The technology is already reshaping science, healthcare, education and economies around the world.
Whether it ultimately narrows inequalities or widens them, and whether it strengthens or weakens democracy and human rights will largely depend on how quickly the world can build governance that keeps pace with innovation.
