TheU.K.government’sminister forartificialintelligence has warned that nations will need to “harden theirdefenses” against AI risk after a former Anthropic employee said that a human extinction-level event could happen in the next decade.
Answering a question fromFortuneon the claim by former Anthropic worker, Jacob Coxon, Kanishka Narayan said he took the warnings seriously. Coxon warned in apost on X, which has been seen 175milliontimes, that “the people building AI earnestly believe that it couldkillus all by the end of the decade.”
Anthropic scientist, Evan Hubinger, agreed with Coxon, saying the risk was greater than 10%. At the weekend Bill Gates, the founder of Microsoft,told NBC’sMeet the Pressthat, in the wrong hands, AI was“certainly powerful enough to drive events that, you know, cause a billion deaths.”
“The first thing to say is that when people in labs tell you that they’re very worried about this, you have to take that seriously,” Narayan said at the annual conference of theLabourParty, the governing party in theU.K. “Youhave toharden and build your defenses. The risks have only grown, so testing is good but clearly insufficient.
“We want to lead the world in being able toconvene[countries] from the G20 next year as part of our presidency. Of course, we want to be fundamental drivers of the architecture of principles that are going to hopefully come out of that.”
U.K.Prime Minister Andy Burnham told the United Nations General Assembly last week that AI frontier labs were a “multiplier of risk” and that the U.K.could play the role ofhonestbrokerbetween competing national interests. Democrats and Republicans have called for new legislation aimed at regulating labs.
U.S.President Donald Trump, who said that talk of AI risks was a hoax, has warned that controlling the pace of US frontier lab research would hand China a significant advantage.
“The risks have only grown, so testing is good but clearly insufficient.”
Kaniskha Narayan, u.k. minister for artificial intelligence
“I have concerns that if wedon’twin AI,we’regoing to be put ina very badposition. We are leading China right now bya pretty goodperiod,” Trump said. The White House is reported to have warned G20 members to take a “hands off” approach to new, global regulations.
Narayan, who sits in the PrimeMinister’s Cabinet, the U.K.’s most powerful political body, said that there werea number ofrisks associated with AI.
“The candid sense I have is that I’m much more focused on risks today in terms of a secular uplift of cybersecurity risks that each of us in our organizations and lives,” he said. “[I am] concerned about the proliferation of these tools in a way thatdrives upbiochemical risk as well, and yes, I am concerned about instances where models go beyond what humans have asked them to do as well.
“The sort of logical loops youhave tojump through to make that an existential questionarequite significant and severe, and sothat’san indicationof how I think about the portfolio of risks here.
“You’vegot to do the best testing in the world to be able to understand the risks and mitigate them.We’redoing that.We’vebeen doing that for three years.We’revery uniquely placed compared to any other country to be able to do that.”
Narayan also said there were many positives from AI,including injob creation, searching for jobs and productivity. He said that theUKhad underplayed the role of innovation and technology IP in driving economic growth.
“We, as organizations, as individuals, as government, as the private sector, we actually have agency in whether this ends up being a technology that augments human labor, or is a technology that automates human labor,” he told a meetingorganizedby the Resolution Foundation, a think tank.
“There are ways of doing technological adoption that means more jobs and more productivity.”
English (United States) 
Chinese