AI Progress Outpaces Science and Policy, UN Panel Warns of Catastrophic Risks
A United Nations independent panel has warned that developments in artificial intelligence are outpacing both scientific understanding and government regulation, leaving no guarantee that the technology will not cause catastrophic harm. The preliminary report by the UN's Independent International Scientific Panel on Artificial Intelligence, released on Wednesday, highlights the growing dilemma policymakers face: they need robust evidence to regulate AI effectively, but such evidence is struggling to keep pace with rapid technological evolution.
“AI capabilities are outpacing both scientific understanding and governments’ ability to adapt,” said Yoshua Bengio, co-chair of the panel, which comprises 40 cross-regional experts. “With growing evidence of deceptive AI behaviour, science currently cannot guarantee that as capabilities continue to increase, AI will not cause catastrophic harm, either on its own or due to malicious users.”
Described as the first global independent assessment of AI’s risks and opportunities, the report aims to provide up-to-date scientific evaluations to help guide decision-making as governments contend with fast-evolving systems. In the near term, it expects a shift towards agentic AI systems capable of carrying out real-world tasks, although growth may be constrained by energy and high-quality data shortages. Over time, it foresees self-improving AI embedded more deeply in the economy and converging with technologies such as quantum computing and biotechnology.
The report notes that AI already demonstrates expert-level reasoning in mathematics and science and is accelerating drug and vaccine development. Its task complexity is doubling every four to seven months, potentially allowing systems to complete work that takes humans days or weeks. While this could deliver significant economic benefits, it remains unclear whether productivity gains from using AI will translate into broader growth or affect jobs.
The panel outlined a range of safety concerns, including the risk of losing control over AI systems as they become increasingly autonomous and deceptive. AI is already being used to generate misinformation and other harmful content and could be exploited for fraud, cyberattacks, and biological threats. Governance remains fragmented, with many countries lacking the capacity to assess or shape advanced AI systems, leaving them reliant on technologies they cannot fully understand or control. Existing safety tools often depend on limited testing data disclosed by companies, the report said.
UN Secretary-General António Guterres urged governments to act swiftly. “The world cannot govern what it cannot understand,” Guterres said in a statement. “The potential is great, but the risks are real, and the cost of waiting is rising.”
In response to growing concerns, global political and tech leaders, alongside the UN’s digital tech agency, announced on Wednesday the creation of a new commission to address AI development. The AI for Good Global Commission will be co-chaired by Rwanda’s President Paul Kagame and Salesforce CEO Marc Benioff, according to a strategic overview document on the commission’s website. ITU Secretary-General Doreen Bogdan-Martin will serve as permanent vice-chair, with other UN agencies expected to participate.
