Japan requests expanded access to Anthropic and OpenAI advanced AI models as part of emergency cybersecurity package

Japan has formally asked Anthropic and OpenAI for greater access to their most advanced AI models, according to a Nikkei report published October 30. The request is part of a broader emergency cybersecurity package in which Japan is seeking to deepen its technology partnerships with the United States.

The push reflects Japan's growing reliance on frontier AI as a tool for national cybersecurity, following a period in which advanced AI models have given both attackers and defenders significantly elevated capabilities. By securing preferential access to cutting-edge models from the two leading US AI laboratories, Japan aims to strengthen its ability to detect and respond to sophisticated cyberattacks.

The move fits into a wider pattern of US-Japan AI cooperation and comes as both Anthropic and OpenAI have been selectively extending access to their most capable models to trusted government-aligned partners for cybersecurity purposes.