Evan Hubinger, who works on the harmony of artificial intelligence with human values at Anthropic, made a statement regarding the post of Jacob Coxon, who resigned from Anthropic, from the social media platform of US-based company X.
“Those who develop artificial intelligence sincerely believe that this technology could kill us all by the end of the decade,” Coxon said. Quoting his post saying, “Jacob is right about this, we truly, sincerely believe that artificial intelligence can kill all humans.” he said.
Hubinger argued that he thought this probability was more than 10 percent in 10 years and said, “I think Anthropic is doing its best, but we do not yet have a plan to solve the problem of compatibility with super intelligence and we do not follow a clear road map in this direction.” made his assessment.
Nobel Prize-winning scientist Geoffrey Hinton told the BBC that people have never produced a product that could be smarter than themselves, so what will happen cannot be predicted.
Hinton stated that assessments that there is a greater than 10 percent probability that artificial intelligence technologies will “kill all humans” within 10 years are not “unreasonable”.
Hinton said that advanced artificial intelligence does not need to be able to control physical systems to pose a serious risk and that “it can cause chaos even by talking to people” and warned that highly capable systems can be used in biological or cyber attacks or in the design of computer viruses.
Hinton emphasized that artificial intelligence should be designed in a way that does not develop targets that will harm people.
