Dodge pass live coding sessions on stage, AI refresher courses, obstacle course gizmos, people walking around with glowing green disco-style headphones blasting UN panel discussions into your ears, and you can take a break. But you might find yourself in the Cyber Zone, on a rotating seating arrangement called UFOTECH that looks more like the kind of lazy Susan you’d meet at a Chinese restaurant than the cyber bench it’s designed to function as.
This is AI for Better Meetingsorganized by the United Nations International Telecommunication Union (ITU), where representatives from the private and public sectors try to discuss how to use technology for the benefit, rather than harm, of humanity.
While Silicon Valley executives and AI lab leaders are testifying to lawmakers in Washington about the dangers of espionage, and the White House is loosening controls on the export of chips, the United Nations Conference on Good Intelligence — now in its 10th year — is focusing on better goals.
“It is our belief that artificial intelligence, deployed responsibly, can help solve humanity’s greatest problems—from hunger to disease to a warming planet,” Doreen Bogdan-Martin, ITU secretary general, said in a keynote speech on the conference’s main stage. “Today, that idea is being tested, along with the challenges that AI itself poses, even as we strive to use it for good.”
What does good mean—and what good does it do to humanity—was the question raised throughout the conference, which spread across a vast 106,000 square meter convention center on the edge of Geneva’s airport district. The hearings were supported by a growing voice of concern that reckless deployment and unregulated corporate monopolies are already undermining international balance and eroding human rights.
For some on the front lines, the utopian veneer of the tech industry has already worn off. Speaking on the sidelines of the event, Giulio Coppi, chief humanitarian officer at campaign group Access Now, called on humanitarian agencies and the public sector to avoid over-reliance on big technology. “We have to be out of the age of innocence,” Coppi says, urging corporations to stop treating tech companies “like your best friends.” He points to a decade of murky, multimillion-dollar deals funded by public money. “You can’t even tell what’s in your technology stack, because it’s constantly changing,” he warns.
Coppi’s opposition was muted compared to some: Pro-Palestinian activists they stormed the stage during a keynote speech by Amazon’s chief technology officer Werner Vogels, claiming that the company’s technology is being used by Israel against the Palestinians, before finally being removed from the hall.
“When we talk about AI, we like the hype, we get excited about it,” says Vijay Janapa Reddi, an engineering professor at Harvard University, during the competition event during the presentation. “Sadly it hasn’t come into practice.” The problem, he says, is that “good” is a very vague standard for an engineer to work against. “When you’re an engineer, beauty means nothing. I can’t build you something beautiful. A plane that flies in five minutes is not beautiful.”
Much of the global debate about AI is now framed around access: Who can use prototypes, who can buy chips, and who is excluded from the computing economy. It’s part of the reason why the Trump administration has implemented, then liftedexport controls on the leading cross-border AI models, and China is located is reported to reflect to make his weight models clear. Enhancing access and cutting off poor countries can leave them dependent on foreign infrastructure platforms and standards.
In a session on AI hardware and the widening digital divide, speakers argued that computing is no longer a technology problem, but a development problem. “If we mean AI for good, and we mean inclusiveness, we have to realize that this is (about) infrastructure development, not just technology,” says Syed Munir Khasru, chairman of the Institute for Policy, Advocacy, and Governance. Others said most major language models remain English-based models, making small, in-house LLMs using cheap hardware necessary if AI is to serve communities beyond the wealthiest markets.
