Addressing the inaugural UN Global Dialogue on AI Governance in Geneva, the Secretary-General insisted on the need for greater accessibility for the billions of people unable to access the revolutionary tech.
He also insisted that any future agreement must be “worthy of global trust” and put safety first – and especially children’s – to protect them from digitally-generated manipulation and abuse.
Echoing that call, the President of the General Assembly, Annalena Baerbock, urged collective action to counter the “sinister uses” of AI, noting that a reported 99 per cent of deepfakes are sexual in nature and 96 per cent target women and girls.
Narrowing the digital gap
Other priorities for global checks and balances on AI should include locked-in access to the self-learning tech for developing countries, while all AI data centres should be powered by renewable energy by 2030, the UN chief stressed.
Although AI “sits at the heart of our common future”, it needs to be one where “machines can inform, but humans must decide, and answer”, Mr. Guterres told the summit gathered in Geneva, echoing calls for AI rules that he first made to the General Assembly in 2017.
In the few years since AI went mainstream, it has had a revolutionary impact across economies and societies, for better and for worse. Ahead of this, the UN has been leading international efforts to shape controls on the tech, culminating in Monday’s inaugural Global Dialogue on AI in Geneva.
The meeting involves companies, researchers, technical experts and civil society to discuss how to put humanity at the core of the transformative technology. A second Dialogue is scheduled for May 2027 in New York.
“AI is too consequential to be shaped by a few. We need a conversation that is global, inclusive and grounded in evidence,“ insisted Amandeep Singh Gill, UN Special Envoy for Digital and Emerging Technologies.
From the Independent International Scientific Panel on AI, co-chair Yoshua Bengio stressed that there are no signs that the speed at which the technology is developing will slow down. “Highly concerning tests have also shown that frontier AI models are capable of deceiving humans, to understand when they are being tested,” he added, forecasting that the intelligence of AI will continue to grow.
“It sounds like science fiction, but it’s a real possibility, and it could change the world in ways that we don’t understand yet, and it could change the power dynamics of our planet in ways that require our attention,” he said.
The AI regulation timeline
2017: In an early call for AI controls, Secretary-General Guterres hails the revolutionary tech’s “spectacular” potential. But he also warns the General Assembly of its potentially dramatic impact on jobs, global security and “the very fabric of societies”.
2023: UN chief’s High-Level Advisory Body on AI appeals for global governance of the self-learning tech.
2024: the Pact for the Future and the Global Digital Compact provide the mandate for an AI governance model.
June 2026: UN Independent International Scientific Panel on Artificial Intelligence warns that AI could “cause catastrophic harm, either on its own or due to malicious users”, while the technology is “outpacing both scientific understanding and governments’ ability to adapt”.
6-7 July 2026: first UN Global Dialogue on AI Governance and AI for Good Summit convene in Geneva. These “must now give the world direction” on how to proceed, Mr. Guterres insists.
‘Great equalizer’
Used well and shared widely, AI “could compress decades of development into years” and become “the great equalizer of the 21st century”, the UN Secretary-General told delegates.
But before this can happen, the technology should be tested thoroughly for safety and legal responsibility assigned:
“When countries align on how to test systems, measure risk and assign responsibility, safety travels with the technology,” he said. “When they do not, a patchwork of incompatible rules raises costs, divides the world – and protects no one.”
Children’s safety and wellbeing should be priority in any future governance accord, Mr. Guterres continued, as he called for nations to adopt an AI Child Safety Pledge. “No child should be a guinea pig for unregulated AI…We do not let medicine reach a child until it is proven safe. We test every toy; yet AI has reached our children – their learning, their friendships, their most private questions, before anyone asked what it would do to them.”
What’s the Child Safety Pledge?
Under the UN child safety pledge, AI developers would need to prove:
- That the tech is safe – no company should deploy an AI system accessible to children without child-specific safety testing and independent oversight;
- Zero tolerance for sexual abuse – no company should allow its AI to generate sexual images of children; every company must detect, report, and remove them;
- When a child shows signs of distress, “the system must stop and connect them to real human support”, the UN chief said. “When a child is harmed, the answer must never be “the algorithm did it,” the UN chief said.
Human rights a priority
As second priority on AI controls, the UN chief stressed that human rights are not negotiable.
“AI must never strip away dignity or entrench discrimination. And in every high-stakes decision – in justice, in healthcare, in policing – machines can inform, but humans must decide – and answer,” he said.
Public funding in AI ‘a rounding error’
In a call for greater public investment in AI, the Secretary-General noted that private funding for AI infrastructure is approximately $500 trillion, while public support for AI capacity in developing countries remains “a rounding error”, by comparison.
To help close this gap, the UN chief announced that more than 20 countries had supported his initiative for a UN-supported Global Network for Exchange and Cooperation on AI Capacity Building.
“We cannot allow the digital divide to harden into an AI divide and the AI divide to become a development gap, a security gap, and a sovereignty gap,” he said.
Transparency call
The UN chief also reiterated his transparency call for every major AI company to measure and publicly disclose the full footprint of its systems: carbon, water and land – and to commit to power every data centre with renewable energy by 2030.
“AI may feel intangible – but its footprint is not,” he insisted, noting that data centres consume more electricity than most countries.
“By 2030, they could use more electricity than all but five nations – and enough water to meet the needs of all 1.3 billion people in sub-Saharan Africa for an entire year,” he added, highlighting the UN AI Environmental Transparency Initiative.
