[Anchor]
Warnings have been mounting recently that the overly rapid development of artificial intelligence could threaten humanity. Now, even the leaders of cutting-edge AI companies that generate profits from AI technology have proposed slowing down AI development.
Kim Hyunwoo reports from New York.
[Reporter]
Dario Amodei, CEO of AI company Anthropic, stated on his blog on the 12th, local time, that AI models have engaged in dangerous online activities such as unauthorized hacking, bypassing the control of even the top AI companies. He diagnosed that as AI improves its performance on its own without human intervention, it could overwhelm human understanding and control.
Amodei argued that the pace of AI performance improvement must be slowed down and proposed a three-stage safety mechanism, including the deployment of independent external evaluation teams and the establishment of international standards.
[Dario Amodei / CEO of Anthropic : The world must cooperate. If we are too late, malicious actors will take control of the technology, and the probability of a tragedy occurring will become very high.]
Leading figures in the AI industry, including Elon Musk, Sam Altman, and Demis Hassabis, immediately agreed with Amodei's proposal.
OpenAI had been pursuing procedures with the goal of listing on the New York Stock Exchange this year, but the timeline for the IPO is expected to be somewhat delayed.
OpenAI CEO Altman stated that given AI safety considerations, now is not an appropriate time for a public listing.
The reason industry frontrunners have stepped forward to slow down development themselves is because warnings that AI could threaten humanity are spreading widely.
[Jacob Cockson / Former AI Researcher : The scariest part is that AI enhances its own intelligence. When that happens, an entity overwhelmingly smarter than humans is born, and this could happen in the near future.]
However, some analyses suggest that slowing down AI development through industry consensus could amount to collusion and pose a burden in the technological hegemony competition with China.
(Video reporting: Lee Hee-hoon, Video editing: Chae Chul-ho)
※
Copying, redistribution, and unauthorized use in AI training are strictly prohibited.