A United Nations scientific panel has issued a stark warning about AI. It says developments are outpacing scientific understanding and government policy. As a result, there are no guarantees the technology will avoid catastrophic harm.
The warning came in a preliminary report on Wednesday. The UN’s Independent International Scientific Panel on Artificial Intelligence produced it. The 40-member group draws experts from across different regions. It calls this the first global independent assessment of AI risks and opportunities.
Policymakers now face a growing dilemma, the panel said. They need robust evidence to regulate AI effectively. Yet that evidence struggles to keep pace with the technology’s rapid evolution. That gap leaves governments reacting to systems they barely understand.
Panel co-chair Yoshua Bengio framed the danger clearly. He said AI capabilities are outpacing science and governments alike. He cited growing evidence of deceptive AI behavior. He warned science cannot yet rule out catastrophic harm from misuse or the systems themselves.
“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.”
The report tracks a striking pace of progress. AI already shows expert-level reasoning in mathematics and science. It is accelerating drug and vaccine development too. Task complexity is doubling every four to seven months. Systems may soon handle work that takes humans days or weeks.
The panel also flagged serious safety concerns. These include losing control over increasingly autonomous systems. AI is already used to generate misinformation and harmful content. It could also be exploited for fraud, cyberattacks, and biological threats.
Governance remains fragmented across nations, the report warned. Many countries lack the capacity to assess or shape advanced AI. That leaves them reliant on technologies they cannot fully control. UN Secretary-General Antonio Guterres urged swift action. He said the world cannot govern what it cannot understand. Leaders also announced a new AI for Good Global Commission to address the risks.
