The UN Security Council was told to “take AI seriously” when it met for the first time to discuss the risks and opportunities of AI.
At the first meeting of its kind at the world’s top diplomatic institution, representatives from 15 member states heard artificial intelligence It brings “catastrophic risks” to mankind, but it is also a “historic opportunity”.
chaired by foreign secretary James CleverlyAt the meeting, the Chamber of Commerce listened to the speeches of Jack Clark, co-founder of Anthropic, a leading artificial intelligence company, and Professor Zeng Yi, co-director of the China-UK AI Ethics and Governance Research Center.
“There is no country that will not be affected by artificial intelligence, so we must engage the broadest coalition of international actors from all sectors,” Mr Cleverly said.
“Our shared goal is to consider the risks of AI and decide how to mitigate them through coordinated action.”
Council members heard from two experts who outlined what they see as the enormous opportunities AI presents, but also pointed to the serious risks it poses and the urgent need for global solidarity on the issue.
“We cannot leave the development of AI entirely in the hands of private sector players,” Mr Clarke said. “Governments around the world must work together to develop national capabilities to further develop robust AI systems.”
“An AI system that can help us understand the science of biology could also be an AI system that can be used to build biological weapons,” he explained.
He warns of the risks of not understanding the technology: “It’s as if we were building an engine without understanding the science of combustion.
“This means that once AI systems are developed and deployed, people will find new uses for them, unexpected by the developers, many of which will be positive, but some of which may be misused.”
Professor Zeng Yi, Director of the Institute The Brain-Inspired Cognitive Intelligence Lab calls on the United Nations as a whole to take the subject seriously.
“The United Nations must play a central role in establishing a framework for AI for development and governance to ensure global peace and security,” the professor said.
“AI is putting humanity at risk of extinction simply because we haven’t found a way to protect ourselves from AI exploiting human weaknesses,” he warned.
Mr Clarke added: “More challenging are the issues of chaotic or unpredictable behaviour. AI systems, once deployed, may exhibit subtle issues that were not detected during development.
“I want to challenge those listening to this talk not to think of AI as a specific technology, but as human labor that can be bought and sold at the speed of a computer, and that will change over time. cheaper and more capable.”
Analysis: Transparency crucial to how AI is trained – regulators must lead
Tony Blair: The impact of artificial intelligence is comparable to the industrial revolution
Microsoft boss says AI ‘not equipped to take over’
‘Huge questions’ about artificial intelligence
Two witnesses raised specific unanswered questions that they said needed urgent answers. Who should gain the power of artificial intelligence? How should the government regulate this power? Which players should be able to create and sell these so-called AI experts? What kind of specialists can we allow to be created?
“These are huge problems,” Mr Clark said. “Humans are rigorously assessed and tested on the job for many key roles.”
Professor Zeng Yi added: “It is very interesting, misleading and irresponsible for a dialogue system driven by generative artificial intelligence to always argue about ‘I think, I suggest’.”
“Well,” he said, “there is no ‘I’ or ‘me’ in AI models. AI should never impersonate a human, take a human position, or mislead humans into false perceptions. We should be using generative AI that can help , but never trust them to replace human decision-making.”
UK to host global artificial intelligence summit
The UK is seeking to play a global leadership role in driving an international consensus on how to use the opportunities to manage the risks of AI. Last month, British Prime Minister Rishi Sunak announced that the UK would host the first major global summit on AI safety in the autumn.
As a permanent member of the Security Council, China welcomes this gathering.
In brief comments to Sky News ahead of the meeting, Ambassador Zhang Jun said he “welcomes the meeting” which will “help to improve understanding of the issue”.
“Let us work together to ensure peace and security as we cross the threshold of a strange world,” Mr Cleverley said at the end of the meeting.