Japan confronts the “dark side” of AI

Japan detailed a global framework for international cooperation on artificial intelligence on May 1, building off the Hiroshima Process announced at last year’s G7 summit. The framework focuses on stemming the risks that generative AI poses for global disinformation — but details are a bit lacking.

“Generative AI has the potential to be a vital tool to further enrich the world,” Prime Minister Fumio Kishida said in an address to the OECD in Paris. “[But] we must also confront the dark side of AI, such as the risk of disinformation.” Kishida said 49 countries and regions have signed on to the agreement, which includes guiding principles and a code of conduct for generative AI developers. It aims to improve transparency, risk mitigation, safety, and authentication.

Eurasia Group analyst David Boling says that the Hiroshima process is a rather incremental attempt at global regulation. “[It] shows that Japan wants to influence how countries regulate AI,” Boling said. “But its approach will be to nudge countries, not judge countries.”

It will be hard for the Hiroshima Process, he said, to “produce anything but statements of high-level principles with limited value.”