Anthropic Expands in Singapore: 1,500 Government Staff Adopt Claude Code

Anthropic is expanding its presence in Southeast Asia as Singapore emerges as an important market for enterprise AI.

On Oct. 14, the Claude developer announced the opening of its fifth Asia-Pacific office and a memorandum of understanding with Singapore's Ministry of Digital Development and Information. The agreement outlines cooperation on AI safety, security, model evaluation, research, and responsible adoption across government agencies and businesses.

The expansion comes as Claude gains traction among Singapore's major organizations, including DBS, Grab, and government technology teams. It also highlights how AI developers are strengthening relationships with regulators and enterprise customers as demand for advanced models grows across the region.

Claude adoption grows across Singapore's government, businesses

Anthropic's new Singapore office will support its growing regional customer base and strengthen relationships with local developers, businesses, and policymakers.

The company said more than 1,500 engineers and technical staff at the Government Technology Agency of Singapore (GovTech) have adopted Claude Code to modernize government systems and accelerate software development.

Singapore also ranks second among 121 countries in Claude.ai usage per capita, according to Anthropic's Economic Index. The company said usage is nearly six times the level its population alone would predict.

Enterprise deployments are extending beyond government agencies. Anthropic identified several major Singapore-based organizations using Claude for software development, productivity, and customer-facing applications.

The deployments illustrate the range of applications Anthropic is targeting, from government software development to financial services and customer support.

The productivity figures are company-reported and should not be interpreted as independently verified results or guarantees of similar outcomes elsewhere.

Anthropic is also building its local developer community. The company said its Oct. 14 Builder Summit Singapore brought together more than 300 developers, while a separate community event earlier in the week attracted more than 700 participants.

Singapore AI safety agreement targets security, testing, and enterprise adoption

Anthropic's agreement with Singapore's government extends beyond the opening of a regional office.

The memorandum, signed on the sidelines of Singapore International Cyber Week, identifies three principal areas for cooperation:

• AI safety, security, and model evaluation: Technical exchanges with Singapore's AI safety and assurance institutions involving model evaluations, testing standards, and incident response.
• Research into AI adoption and its impact: Sharing research and data on how AI is being adopted and its effects on businesses, workers, and the wider economy.
• Responsible AI adoption: Exploring ways to help public agencies and private businesses, including smaller organizations, deploy advanced AI systems responsibly.

Alongside these areas, Anthropic plans to support AI skills development, help build AI capabilities among public officers, and offer technical and policy expertise as Singapore develops its approach to AI governance.

The agreement builds on Anthropic's earlier involvement in Singapore's Generative AI Evaluation Sandbox, which began in 2023. The company subsequently participated in multilingual AI red-teaming involving English, Mandarin, Malay, and Tamil.

The memorandum outlines areas for future cooperation. It does not, by itself, establish new mandatory technical standards for organizations using Claude or give Singapore priority access to Anthropic's frontier models before their commercial release.

According to Channel News Asia, Singapore has also reached AI cooperation agreements with OpenAI and Microsoft in 2026, reflecting the government's broader efforts to develop AI capabilities while addressing security and safety concerns.

For regulated industries, infrastructure is another consideration. Anthropic says organizations can access Claude through Amazon Bedrock with in-region inference, potentially helping businesses address strict data-processing requirements. Organizations must still verify the availability of their required models and applicable compliance controls.

The expansion also has limits. According to The Business Times, Anthropic is working to bring its most advanced models into Singapore-based infrastructure but has not announced a timeline. The company also has no immediate plans to establish a core AI research operation in Singapore.

What eWeek found: Singapore is becoming an enterprise AI proving ground

Anthropic's Singapore expansion illustrates three developments that could shape enterprise AI adoption across Southeast Asia:

• Enterprise adoption is moving beyond experimentation. Banks, government agencies, and major technology companies are integrating Claude into software development, customer service, and internal operations.
• AI safety is becoming a competitive consideration. Businesses operating in regulated industries must evaluate model capabilities alongside security controls, data-processing arrangements, and compliance requirements.
• Singapore could influence wider regional adoption. Deployments involving organizations operating across Southeast Asia may offer lessons for other markets, although local regulations and infrastructure requirements will differ.

For enterprise technology leaders, Anthropic's expansion raises practical questions about how to evaluate AI suppliers.

Organizations considering Claude or competing systems should examine where data is processed, how model performance is evaluated, what safeguards are available, and whether vendors can support local implementation requirements.

The reported productivity gains at organizations such as Grab illustrate potential benefits, but those outcomes should not be treated as guarantees for other deployments.

Anthropic has not disclosed a specific financial investment or regional hiring target associated with the office opening, although the company says its Singapore office is hiring.

The next test will be whether closer relationships with businesses and government institutions translate into measurable improvements in AI reliability, security, and operational performance.

As AI companies expand across Southeast Asia, those results may matter more to enterprise customers than the number of new offices they open.

More APAC news: For more on Singapore's growing role in enterprise AI, read how Databricks is investing more than $350 million in the country to help businesses move AI applications from experimentation into production.