Anthropic suggests a global pause in AI development
The company is advocating to "temporarily" halt development to prevent artificial intelligence from escaping human control.
Anthropic. File archive
(AFP) Artificial intelligence company Anthropic on Thursday proposed a global pause in the development of A.I. in the face of indications that newer models might escape human control.
The firm, based in San Francisco and developer of the Claude tool, said in a report that a worldwide slowdown in next-generation A.I. development would "likely be a good thing."
However, it warned that if just one company slows down, rivals could simply jump ahead in the race.
"We believe it would be good for the world to have the option to slow or temporarily pause frontier A.I. development to enable societal structures and alignment research to keep up with the advance of the technology," the company said.
An actual suspension of work would mean multiple A.I. giants in various countries, primarily in China and the United States, agreeing to stop unanimously, under rules that everyone can verify, Anthropic said.
"Without a global coordination mechanism, companies and governments will have to make difficult decisions about safety while under competitive and geopolitical pressures," it said.
Other industry players and White House officials have rejected the company's position, saying it is focused on worst-case scenarios and exaggerates risks to thwart rivals under the guise of security.
Still, the White House has acknowledged the power of the company's Mythos model, which has not been made available to the general public because of its cybersecurity capabilities and is currently only used in a small number of select organizations.
American officials and big tech executives have further argued that slowing A.I. development could cede a significant advantage to China.
However, President Donald Trump said that his recent visit to Beijing discussed the possibility of cooperating with the country on A.I. security issues.
Trump also signed an executive order this week that will allow the government to conduct preliminary evaluations of the most powerful A.I. models from American companies before they are launched.
LATEST
