Anthropic has opened an office in Singapore, its fifth in Asia-Pacific, and signed a memorandum of understanding with the government. Neither comes with a figure or a headcount.
The memorandum covers AI safety, security and model evaluation, research on adoption, and responsible adoption across the public and private sectors, the company said.
Co-founder Ben Mann signed it with Kai Fong Chng, permanent secretary at the digital ministry, on the sidelines of Singapore International Cyber Week. Minister Josephine Teo witnessed it.
The plans are technical exchanges with Singapore’s safety and assurance institutions on model evaluation, testing standards and incident response. Anthropic also offered skills work and Economic Index data.
OpenAI put money behind its own Singapore push. Its first applied AI lab outside America arrived in May with S$300M, about $235M, and roughly 200 roles.
Anthropic’s numbers are about usage. More than 1,500 GovTech engineers use Claude Code, Grab says its Claude sales assistant saves about 40,000 hours a quarter, and DBS and GIC are rolling it out.
The jurisdiction matters more than the ceremony. Our own July reporting found OpenAI and Google selling advanced AI to Singapore-registered arms of Alibaba, Baidu and Tencent, whose parents sit on a Pentagon list.
US export controls name entities and places, and Singapore is not restricted. Anthropic bars Chinese companies and their foreign-owned entities from its frontier models, and has pushed Washington for broader controls.
Europe asks for the same cooperation by law. The AI Office can demand documentation and request access to frontier models, and it has struggled to get it, including for Anthropic’s Mythos.
Anthropic signed up there. It agreed to the EU’s code of practice for general-purpose AI in July 2025, and has asked governments for power to block deployments posing a significant risk of catastrophic harm.
What evaluators may publish is unsettled. Reuters reported on Tuesday that Anthropic and OpenAI have sought control over what reviewers can make public, citing trade secrets. Anthropic says it keeps sign-off only with new firms.
Every number in the announcement is Anthropic’s own. That includes the claim that Singapore ranks second of 121 countries on its Economic Index, which measures Claude use rather than AI adoption.
The office is open and hiring, with no target given. In Dublin the same day, Anthropic’s head of international said open models carry more risk by definition.