UN AI for Good summit debates who controls AI access and rules

In short: The UN’s AI for Good summit highlighted growing concern that AI is moving faster than global rules and fair access can keep up.

What's going on

The United Nations’ International Telecommunication Union (ITU) held its 10th annual AI for Good summit in Geneva. The event mixed flashy demos, like humanoid robots and big tech booths, with serious discussions about how AI should be used and who gets to shape it.

In a keynote, ITU secretary-general Doreen Bogdan-Martin said the goal is to use AI responsibly to help with major problems like hunger, disease, and climate change. But many sessions returned to a hard question, what does “good” actually mean, and how do you measure it.

Some speakers warned that governments and aid groups rely too much on large technology companies. Giulio Coppi of Access Now said public money has funded large deals that can be hard to explain or track over time, because the tools and vendors keep changing. Pro-Palestine activists also disrupted a talk by Amazon CTO Werner Vogels, alleging Amazon technology is being used by Israel against Palestinians.

A big theme was access to “compute,” meaning the powerful computers and chips needed to run advanced AI (like needing an expensive engine to drive a heavy truck). Speakers argued this is becoming a development issue, not just a tech issue, and that limits on access can leave poorer countries dependent on foreign platforms. Others noted many AI systems work best in English, which can sideline local languages.

What to watch

The UN is promoting a new 44-member commission, cochaired by Rwanda’s President Paul Kagame and Salesforce CEO Marc Benioff, to guide AI for Good efforts. The next test is whether groups can turn broad promises into enforceable standards and practical checks, before AI products spread faster than global agreement.

Source: Wired