US President Donald Trump and UN Secretary-General António Guterres offered conflicting visions for how the world should manage artificial intelligence at the UN on Tuesday.
Armed with a new UN science report that warns of a slight but real potential of catastrophic and irreversible harm, Guterres kicked off the parade of world leaders speaking at the General Assembly by calling for more oversight of AI, warning in vivid if not ominous terms of what could go wrong.
“Let us resolve that life-and-death decisions must never be surrendered to machines. Killer robots must have no place in our future,” the UN chief said in his farewell address to world leaders. He also warned about AI in his first General Assembly address a decade ago.
With Guterres saying the world needs to “advance the conditions for the responsible pacing of AI and the additional safeguards needed to address risks,” the UN Security Council on Wednesday will address how to place some kind of controls on the technology and it will feature some of the biggest names in the field.
It’s super, not artificial, Trump says.
Less than an hour later, Trump was having none of that.
“The United States also totally rejects any attempt to construct a globalist scheme to control for the artificial intelligence being spoken of so much. Now here and after officially called super intelligence, changing the name,” he said.
World leaders speaking after Trump used “artificial intelligence” — not “super.”
“Every major new technology brings challenges, and superintelligence is no exception,” Trump said. “Yet the very same people who said we’ll all be dead in 12 years because of global warming, a name since reborn to climate change, because the planet was cooling, not warming, and nobody was dead.”
“These are the same people that are now saying that AI is going to kill us all, that robots are going to attack us, and that everything is going to be a total disaster.”
Go slow but cooperate, others say
The problem is both Trump and Guterres reduce the issue to something that’s not useful, says University of Chicago computer science department chairman Henry Hoffmann.
“One turns it into a science-fiction movie; the other turns it into a nonissue,” Hoffmann said in an email.
“Both push people to pick a team instead of engaging with the real risks and the possible mitigations. There shouldn’t be ‘sides’ on whether we need adequate control over powerful autonomous systems. The interesting debate is over what mechanisms actually provide that control.”
Other nations want to make sure they are not left behind during an AI race between the US and China, which Trump said America is winning. French President Emmanuel Macron and Guterres also made the issue about protecting children.
“Let’s not wait for the consequences to come raining down on AI,” Macron said. “We must act now to protect our children and our young people.”
Finland President Alexander Stubb said AI will change the world, potentially for good, but needs strong controls.
“Artificial intelligence is the most consequential technology in the history of mankind. It will impact science, the economy, politics, security and the very essence of our humanity,” Stubb said.
“Humans can and should be in the driver’s seat. We have a choice. We can compete, cooperate or slow down. For me, cooperation is the best way forward.”
British Prime Minister Andy Burnham told a meeting of business leaders on the sidelines of the UN General Assembly that he wants to see “a single set of global principles and standards” that can unlock AI’s potential while mitigating its risks.
He said forging “AI for good” will be the main focus of the UK’s presidency of the G20 group of nations next year.
At Wednesday’s Security Council AI meeting, the UN is hearing from the people in building these AI systems, as well as those who have recently sounded the alarm about the need for guardrails.
Taking the podium are Sam Altman, the CEO of OpenAI, followed by Dario Amodei, the CEO of Anthropic, and finally Clément Delangue, the CEO of and co-founder of Hugging Face, which was hacked by a rogue OpenAI program.
Additional sources • AP