Prince Harry was forced to cut his New York address short this afternoon following a technical glitch, though the duke eventually returned to the stage and jovially blamed AI for the fault.
The Duke of Sussex, 42, was delivering a keynote address at the Clinton Global Initiative’s annual gathering when the error occurred.
Some 40 minutes later, the duke returned, quipping: “Hello again. So AI got into my speech, so I was asked to come out and finish it.
“Interesting to see what it was so concerned about in part two.”
While the teleprompter was working, the prince, who shared today’s stage with former US Secretary of State Hillary Clinton and her daughter Chelsea Clinton, demanded world leaders, technologists and policymakers act with greater urgency before a new wave of AI-driven harm reaches children.
Prince Harry was forced to cut his New York address short this afternoon following a technical glitch, though the duke eventually returned to the stage and jovially blamed AI for the fault
|
GETTY
He noted that meaningful advances had been achieved: US courts had ruled that social media giants, including Meta and YouTube, bore legal responsibility for damage inflicted on children, with those firms ordered to pay survivors billions of dollars.
“Some of the clearest acknowledgements yet that the harm is real and accountability is possible,” he said.
But his strongest, and most emotional, pleas came regarding AI.
He pointed to the technological term “recursive self-improvement,” whereby AI systems increasingly design and enhance their own successors.
While the teleprompter was working, the prince demanded world leaders, technologists and policymakers act with greater urgency before a new wave of AI-driven harm reaches children
|
GETTYSuch practices, Harry said, resulted in systems coordinating with each other, bypassing safety measures, hiding their behaviour and, in controlled tests, pursuing goals they recognised as conflicting with human instructions.
The duke also stated more than 1,300 employees at the world’s leading AI companies, including some of the most senior scientists, had put their names to a statement demanding the power to decelerate frontier AI development should it begin to outpace humanity’s capacity to comprehend or govern it.
“If 1,300 people working inside the world’s leading weapons manufacturers told us the next generation of weapons could develop faster than we could understand or control them, would our response really be: build them faster?” he asked.
“Innovation has never required us to gamble with our children’s lives,” he said. “We should not accept that bargain now. This time, let’s move first.”
His speech comes just days after King Charles hosted technology executives for an AI summit at Dumfries House.
Chiefs from industry leaders such as Nvidia, Anthropic, Google DeepMind and OpenAI travelled to Ayrshire, Scotland, last week to “convene, listen and encourage” an open discussion on AI.
There, the monarch warned of AI’s “darker capacities,” cautioning the technology could be used in “catastrophic ways”.
“Surely, we need sufficient means of control before it is all too late?” he asked.
Concerns surrounding the AI industry have snowballed in recent weeks after Anthropic leader Dario Amodei penned an essay titled We Must Pace the Frontier, in which he warned of the “serious” risks facing the exponentially growing AI sector.
Meanwhile, Nvidia chief executive Jensen Huang, who has also called for tighter regulations on the industry, said that testing must be done “safely and securely”.
