Photo/Illutration Sam Altman, CEO of OpenAI, responds in an interview with The Asahi Shimbun in Tokyo on June 12. (Wataru Sekita)

The developer of the ChatGPT generative artificial intelligence service is stepping up efforts to open an office in Japan, a country he said he highly values for its technological advancements.

In an exclusive interview with The Asahi Shimbun in Tokyo on June 12, Sam Altman, CEO of OpenAI, said he had hired James Kondo, who once headed Twitter’s Japan arm, to lead its Tokyo base.

But he did not reveal when the office would be open for business.

Altman was in Tokyo as part of a global tour that includes stops in 22 nations. Before the tour, Altman visited Tokyo in April for a meeting with Prime Minister Fumio Kishida at the prime minister's office.

Japan was the first overseas destination for Altman since his company began providing the ChatGPT service.

When asked why his company placed such emphasis on Japan, Altman said, “Japan has always been a country at the forefront of humans and machines working together.”

He added, “We've got many great partnerships in progress with Japanese companies.”

Altman has met with many national leaders during his global tour and has discussed what sort of regulations were needed.

“Leaders are aligned around the world about thinking about global regulation and how to balance the tremendous upsides of this (technology) without causing real problems in the short and the long term,” Altman said.

He also talked about the benefits and risks involved with the most advanced generative AI technology such as ChatGPT.

"This is technology that I think will do far more good for the world,” he said. “The benefits are going to keep getting bigger and bigger. But along with that the risks get commensurately bigger, too.”

In the medium term, he mentioned the possibility of abusing the technology for bioterrorism attacks, while a long-term concern was its abuse by authoritarian states.

“I think it more likely to be a tool used in war in a way that could be quite dangerous,” Altman added.

He said that generative AI technology significantly differed from conventional AI.

“The main thing that is interesting about this new wave is not necessarily the 'generative-ness,' although people like that too. But it's the 'general-iness,'” Altman said. “Computer programmers we talked to said, ‘I don't think we've seen a technology like this in a while now.’”

He added that the mission of his company was to bring about artificial general intelligence (AGI) that would be beneficial to everyone.

“I think AGI is at this point a very poorly defined term,” Altman said. “A definition that we like to use is that it can contribute to half of the economically valuable current jobs. By 2030, we will have extremely powerful AI systems. I think they'll have a significant impact on human society.”

While Europe has been a region that has moved the fastest in placing regulations on generative AI, Altman said that in his meetings with European leaders, “Views are different in different places, but generally speaking, they really want to make sure that Europe benefits from this technology.”

Altman said he participated in an online meeting with Chinese officials two days ago and planned to hold further discussions in the future.

Altman also touched upon the company’s decision to come out with an application programming interface (API) that developers outside the company could use.

“One of the reasons we wanted to release the API was to decentralize the power,” he said. “This could have been locked up in a few companies, but now anybody can use these models.”

Related News