Recently I sat in on a meeting between an AI infrastructure founder and a senior director at Singapore's Economic Development Board (EDB). The company runs a multimodal API platform with ARR approaching $100M, incorporated in Singapore but with a mostly remote team.
What followed was one of the most practically useful hours I've spent in a meeting room. I'm sharing the key takeaways here — stripped of anything company-specific — because this is exactly the kind of information that AI founders preparing for Singapore need and rarely get straight.
How EDB Defines "Singapore Company"
The first question every China-origin founder asks: will I be treated as a "Chinese company" or a "Singapore company"? The answer is more systematic than most people expect.
EDB evaluates three factors: where the company is incorporated, whether the shareholding structure includes Chinese entities, and whether revenue and customers are global. If you're incorporated in Singapore, your cap table is clean of China-domiciled entities, and your business serves international markets, you're a Singapore company. Full stop.
This distinction matters because it determines whether you're on the "headquarters track" or a track that requires more scrutiny. Get your corporate structure right early.
Employment Pass: Simpler Than You Think
The EP process is a constant source of anxiety for founders, but the reality is less dramatic than the rumors suggest.
For core executives — CEO, CTO — the threshold isn't particularly high. You don't need to hire local employees first. Salary must be paid from the Singapore entity. And here's the stat that should calm most founders down: roughly 90% of EP rejections happen because of errors in the application, not because of policy barriers.
The advice from EDB was refreshingly direct: apply online yourself first. If you get rejected, then bring in a specialist. Don't start with an immigration consultant.
Company stage matters: Early-stage companies (10–20 people) face the least friction. As you scale, workforce diversity requirements become more significant.
The Real Red Line: Team Homogeneity
Singapore doesn't enforce strict local hiring quotas for tech companies, but there is one thing the government watches closely: team composition.
A team that is entirely Chinese won't fly. Neither will a team that is entirely Indian, or entirely American. What works is genuine diversity — a mix of nationalities and backgrounds. This isn't about political correctness. It's based on a long-standing observation: homogeneous teams rarely succeed at true internationalization.
If your plan is to transplant your entire China-based engineering team to Singapore, rethink it. A mix of China, North America, Europe, and APAC talent is what EDB wants to see.
Special Intervention Is the Exception, Not the Rule
Every founder has heard stories about certain companies receiving extraordinary government support — EP fast-tracking, direct intervention from senior officials. The meeting clarified something important: those cases involved companies under extreme external regulatory pressure with no alternative options. EDB stepped in because the company had to relocate its core team within weeks, not months.
Normal companies following normal processes have a higher success rate than companies seeking special treatment.
The takeaway is counterintuitive but important: don't try to manufacture urgency. Just go through the standard channels.
The Right Order of Operations
When your company reaches significant scale, most founders want to jump straight to tax incentives. EDB's recommended sequence is different:
Step 1: Tax compliance. Get your international tax structure and transfer pricing right. Be able to explain clearly why revenue flows the way it does and why profit is allocated where it is. Use a Big Four firm, not a boutique.
Step 2: Establish your HQ positioning. Demonstrate that Singapore is your genuine operational center, not a mailbox.
Step 3: Only then explore tax incentives. The standard corporate rate is 17%. With roughly 15 employees, you can negotiate to 15%. With 25 employees by year five, potentially 10%.
Key insight: Singapore's incentive policies reward structures that are already working. They're not designed to fix broken ones. Get compliant first, optimize later.
The High-ROI Programs Most Founders Miss
Rather than chasing headline tax rates, EDB pointed to two programs that deliver outsized value for early-stage AI companies:
R&D grants. Available for technical teams of 10 or fewer. As long as the work is genuine research connected to your core technology, you're eligible. This is real money, not symbolic.
Graduate training subsidies. Hire five local fresh graduates full-time, have your CTO or senior engineers train them, and the government subsidizes a portion of their salary. It's a workforce development program, but for AI companies it's also a way to build local talent while reducing burn.
There's also the Enterprise Compute Initiative, where the government subsidizes local enterprises using AI infrastructure. If you're an AI infra provider with a Singapore presence, this opens up a channel to co-sell alongside AWS, Azure, and GCP.
What This Actually Means
The meeting confirmed something I've observed repeatedly in my work: Singapore's policy environment is designed for companies that have already achieved product-market fit and are now facing structural questions — legal entity design, compliance, long-term sustainability.
If you're still searching for PMF, government engagement isn't the priority. But when the risk shifts from "can we build something people want" to "can we scale without our structure falling apart," these conversations become essential.
This is an installment of Field Notes, where I share first-hand observations from Singapore's AI ecosystem. If you're an AI founder thinking about Singapore, reach out — conversations like these are what I do. Subscribe to Field Notes for more.