Singapore never set out to referee the AI arms race. Yet here it stands. U.S. labs rush to open applied research outposts. Chinese startups quietly register entities and hire local talent. All chase the same promise of distance from Beijing’s oversight and Washington’s export controls.
The trend accelerated this year. Fortune reported June 19 that OpenAI committed S$300 million to the local AI scene and launched its first applied AI lab outside America. Anthropic posted jobs for finance, product support and economic research roles. Google DeepMind set up shop too. On the other side, Tencent deepened its footprint. San Francisco-based Plaud AI, maker of an AI notetaker, pledged S$10 million to expand operations and grow headcount from 100 to 150 by year-end.
“All the AI companies I work with, whether they’re from China, Korea or Japan, all use Singapore as a hub,” Gunja Gargeshwari, chief revenue officer at Israel-based Bright Data, told Fortune on the sidelines of the SuperAI summit. Bright Data itself picked the city-state for its APAC headquarters despite 60% of its Asian customers sitting in China and India.
The logic cuts two ways. Chinese founders want to reassure global clients their code won’t disappear into state demands. U.S. teams hunt talent without battling H-1B lotteries or security clearances. Singapore offers both. Stable rules. English-speaking workforce. Top universities. National University of Singapore sits at No. 8 in the latest QS rankings. Nanyang Technological University ranks 12th.
“We have the chance to stand out here,” Nathan Xu, CEO of Plaud, said in the Fortune piece. The company hired its first local employee in 2024. Now it senses momentum. “Unlike many companies that originate entirely from the U.S., if Plaud can position ourselves aggressively in Singapore, then we’re a cool company to prospective users across the globe.”
Prime Minister Lawrence Wong captured the national brand last July. “Some say we are boring, and we will never have the same offerings as New York and Paris. But at the same time, we are stable, we are predictable. We are reliable and we are trusted, and these are intangible assets that others would die to have.”
Geopolitical Shelter With Visible Cracks
Yet shelter comes with limits. Beijing proved last spring it can see through incorporation papers. Manus AI, founded by Chinese entrepreneurs, moved its global headquarters to Singapore in mid-2025. The move aimed to dodge Western rules and tap international capital. In December the firm agreed to a $2 billion sale to Meta. Chinese authorities blocked it. By April they forced the deal unwound.
Regulators ignored the Singapore flag. They focused on where the technology originated and who built it. “Regulators looked straight through the Singapore holding structure to the technology’s Chinese origin,” Sebastian Wiendieck, head of legal practice in China at law firm ROEDL, told CNA, as cited in Fortune. The episode sent a clear signal. Any China-founded AI startup faces national security scrutiny when selling to U.S. buyers, regardless of domicile.
Washington applies pressure too. Last week the U.S. government barred non-U.S. individuals from using Anthropic’s powerful Mythos model. Singapore risks losing access to frontier systems. Earlier reporting from Reuters on April 24 had already flagged the shift. The city-state moved from East-West gateway to neutral ground. Chinese startups sought freedom from government attention and political bias. American firms chased talent without visa headaches.
“Singapore is increasingly becoming a neutral hub for AI companies from both the U.S. and China,” Brad Gastwirth, global head of research at Circular Technology, told Reuters. Anthropic, whose $30 billion fundraising round was led by Singapore’s GIC sovereign wealth fund, planned a local office. It joined OpenAI, Meta’s Superintelligence Labs and Alphabet’s DeepMind.
The Reuters dispatch noted Chinese firms dangle big pay. PhD-level AI researchers in Singapore can command $150,000 to $273,000 annually. Some researchers cannot leave China without notifying authorities. An office here becomes table stakes for global business.
But. The Manus episode exposed fragility. A Singapore-based company with Chinese roots and U.S. buyers no longer counts as neutral. Investors now weigh origin and control more heavily than registration papers. Regulators on both sides tighten focus.
Still, activity continues. Notion opened a Singapore office in mid-2025. Its head of design, Randy Hunt, explained the draw. “Our number one priority is to meet and interface with current and potential customers. I could do a demo for you over video, and while that may be effective, if I can do it sitting next to you, it resonates better.”
Enterprise demand pulls harder than consumer apps in this market. Many multinationals base APAC operations here. Anthropic’s enterprise focus fits perfectly. So does the broader industry turn. After years of pouring capital into ever-larger models, attention swings toward real returns. BNY wealth analysts noted in a March report that 2025’s defining trait was capital expenditure. Now focus lands on monetization and deployment.
Singapore feeds that shift. It released a national AI R&D plan in January and injected S$1 billion into infrastructure. Plans for Kampong AI, an industrial park with workspaces and housing, target 2028 opening. The goal is clear. Attract startups. Keep talent. Build applied capability.
SuperAI 2026, held in June at Marina Bay Sands, sold out. The conference drew 10,000 attendees from 150 countries. Organizers positioned it as one of the few venues where U.S., Chinese, European and Asian players share the same room. Peter Noszek, co-founder, captured the moment in a May PR Newswire release. “SuperAI exists because the global AI ecosystem needs a place where the US, China, Europe and Asia can meet in one room.”
Recent X discussions echo the momentum. Posts from early June highlighted Anthropic hiring a country lead and support roles. OpenAI’s applied lab and government partnership drew notice. Nvidia and Singtel advance data center plans. Every major frontier lab seems present.
Yet electricity constraints linger. Local grids cannot easily support massive training clusters. Some analysts question long-term scalability for inference at scale. Talent poaching runs hot. Chinese firms reportedly offer Singapore engineers packages around S$280,000 to close gaps back home.
The city-state bets its predictability outweighs these frictions. Founders feel welcomed. Xu of Plaud put it plainly. “We didn’t know we’d be able to set up such a big and meaningful presence here; a year ago, we had zero people here, but now we have close to a hundred.”
That sentiment may prove decisive. As export rules tighten and investment screens sharpen, third locations gain value. Singapore offers regulatory clarity, talent pipelines and geographic reach into Southeast Asia’s growing markets. It cannot replace Silicon Valley or Zhongguancun. It doesn’t need to. It simply needs to stay useful to both sides longer than either expects.
Whether that balance holds through the next wave of model releases and regulatory salvos remains the open question. For now, the offices keep opening. The talent keeps arriving. And the small island nation keeps collecting intangible assets others would die to have.
WebProNews is an iEntry Publication