Anthropic has denied a request from a Chinese think tank for access to its latest AI model, Claude Mythos, which was announced in early April 2026 as a leading technology for defensive cybersecurity. This decision underscores the heightened security concerns surrounding Mythos, which has raised alarms among central banks and intelligence agencies globally due to its advanced capabilities. The rejection also reflects the ongoing tensions in US-China competition over artificial intelligence, with Anthropic enforcing strict controls on its technology’s access.

Anthropic: Anthropic is an AI safety and research company dedicated to building reliable, interpretable, and steerable AI systems that people can rely on, while generating research on AI’s opportunities and risks. As of 2026, it focuses on frontier AI research across modalities, safety techniques, and deploying products like Claude to businesses, nonprofits, and civil society. It recently rejected a request from a Chinese think tank for access to its newest AI model during talks in Singapore.
New York Times: The New York Times is a leading American newspaper renowned for its in-depth reporting on technology, politics, and global affairs. It has provided extensive coverage of AI developments, including Anthropic’s advanced models like Mythos. The outlet reported on Anthropic’s refusal to grant a Chinese think tank access to the latest AI model.

`json
{
“Latest Model”: “Anthropic’s latest AI model is designed for advanced applications, with strict access controls.”,
“US-China Dynamics”: “The rejection highlights ongoing tensions in US-China AI competition, with Anthropic maintaining strict controls on access to its technology.”
}
`