The UN states that AI is advancing more rapidly than regulations, and it has a report to back this claim.

      The United Nations has compiled several of its concerns regarding artificial intelligence into a single document, with a clear main finding: AI capabilities are progressing at a rate that surpasses any government’s ability to comprehend, evaluate, or regulate them. This warning coincides with the gathering of delegates in Geneva for the UN Global Dialogue on AI Governance and comes amid a policy environment where the EU's AI Act is one of the few binding frameworks in effect.

      The warning stems from a preliminary report by the UN’s Independent International Scientific Panel on AI, released on July 1, which is presented as the first comprehensive global evaluation of the technology. Its main assertion highlights a gap between the current abilities of AI systems and the scientific knowledge necessary to govern them effectively. The panel contends that regulation is lagging, and so is the fundamental research required for policymakers to create sound regulations.

      Secretary-General António Guterres articulated the message more simply, stating, “The more AI advances without shared rules, the less influence governments and people will have on the outcomes.” He gave a two-word piece of advice to governments: “Do not wait.” He reiterated concerns about understanding multiple times, emphasizing that “the world cannot govern what it cannot understand,” and warned that while the potential is significant, the risks are substantial, and the cost of inaction continues to grow.

      The report's strength lies in its framing of governance struggling to keep pace with a rapidly evolving object that it cannot yet quantify. Rather than focusing on a single catastrophic scenario, it highlights a structural mismatch where the speed of technological advancements outstrips the slower processes of evaluation, standard-setting, and legislation—a common concern shared by researchers in AI governance.

      An immediate counterargument is that governments are taking action. The EU has implemented a risk-based regulatory framework, despite inconsistent application among member states. China has moved to limit humanlike AI agents, necessitating adjustments to consumer products already available. Conversely, the United States has faced challenges in establishing cohesive federal regulations, resulting in a regulatory void that critics argue leaves the nation ill-equipped to oversee the industry it hosts.

      The panel asserts that these initiatives are fragmented, and such fragmentation poses its own risks. An equity issue is also woven throughout the assessment, with experts warning that the opportunity to shape AI is dwindling. If this opportunity slips away, leading to technology being concentrated within a few companies and nations, it could exacerbate global inequality rather than alleviate it. The distribution of access to computing, data, and talent, as well as the ability to govern, is uneven.

      The report does not specify a particular institution or treaty but instead contributes to the Geneva dialogue, which aims to initiate a process rather than reach a definitive conclusion. The UN has intentionally positioned the panel as an advisory body, somewhat akin to the scientific assessments that guide intergovernmental discussions without dictating outcomes.

      Whether this approach can advance at the pace described in the report remains uncertain. Intergovernmental processes are inherently slow, and the panel’s main conclusion is that AI is not. The climate analogy is relevant from both perspectives: while reports have created a shared evidence base, decades of assessments have not ensured decisive action. The panel is hopeful that establishing a common scientific foundation remains beneficial, even if political responses lag behind it.