AI's $50 Billion Week — What It Means for Singapore Businesses
If you only read one tech story this week, here's the headline that sums it up:
“<em>Google committed up to $40 billion to Anthropic. Thinking Machines Lab signed a multi-billion dollar deal with Google Cloud. Sea Limited opened a new AI Centre of Excellence right here in Singapore. All in seven days.</em>”
That's roughly S$60 billion in AI infrastructure deals announced between April 20 and April 26 — a single week. And while these are foreign-sounding names making foreign-sounding announcements, the second-order effects are landing right at our doorstep.
Here's what happened, why it matters, and what your business should actually do about it.
The Big Three
1. Google ↔ Anthropic: Up to $40B
On April 24, Google announced it would invest up to $40 billion into Anthropic — $10B immediately, with the remaining $30B tied to performance milestones. The deal includes massive access to Google's tensor processing units (TPUs).
Translation: Anthropic gets the compute it needs to keep training frontier models. Google gets a hedge against losing the AI race to OpenAI. And the rest of us get cheaper, faster, more capable AI models in the months ahead.
2. Thinking Machines Lab ↔ Google Cloud
On April 22, Mira Murati's new lab — founded by ex-OpenAI executives — locked in a multi-billion dollar deal with Google Cloud. It's the lab's first cloud partnership and gives them access to Nvidia's latest GB300 chips for training their next models.
Why this matters: more well-funded labs at the frontier means more competition, which means more choice and lower prices for businesses downstream.
3. Google's New TPUs
Google Cloud unveiled its newest generation of TPUs the same week. These are the chips that compete directly with Nvidia for the AI training and inference market. Cheaper compute upstream = cheaper AI everywhere downstream.
Closer to Home: Singapore's AI Week
While Big Tech was making global headlines, Singapore was quietly making moves of its own.
- Sea Limited announced an AI Centre of Excellence in Singapore on April 20, with plans to create at least 100 R&D roles over the next three years.
- Singtel's RE:AI, the sovereign AI cloud business of Singtel Digital InfraCo, signed a strategic partnership with Mistral AI to scale AI adoption across financial services, defence, government shared services, and healthcare.
- Microsoft is well into rollout of its $5.5 billion Singapore AI investment announced earlier this month — including free Microsoft 365 Premium with Copilot for every tertiary student in the country.
- A new PwC Global AI study released this month found that 67% of Singapore businesses have a higher risk appetite for AI investment — compared to just 41% globally.
And in a separate analysis published this week, Singapore is increasingly being described as "neutral ground" for the AI sector — Chinese AI startups setting up here to operate beyond geopolitical reach, U.S. firms recruiting talent from the region. Both sides are choosing Singapore. That has implications for hiring, partnerships, and the kinds of vendors you'll be able to choose from over the next 24 months.
What This Means for Your Business
If you're running an SME in Singapore, three takeaways from this week:
1. AI is getting cheaper, not more expensive
Every billion-dollar deal between cloud providers and AI labs ultimately drives down the unit price you pay per query. Models that cost dollars to run a year ago now cost cents. The "wait and see" tax on AI adoption is shrinking — but the competitive cost of waiting isn't.
2. Your competitors are moving — even if you can't see it yet
When 67% of Singapore businesses have a high risk appetite for AI, that's not just trend data. That's the company across the road quietly automating their reporting, their customer enquiries, their inventory monitoring. By the time it's visible from the outside, it's already a year of compound advantage.
3. The talent pool is growing — use it
Sea's 100 new R&D roles. Microsoft's free Copilot for every tertiary student. Singtel-Mistral's enterprise partnerships. Singapore is deliberately becoming an AI hub, and the people and tooling to deploy AI in your business have never been more accessible. The bottleneck isn't talent anymore — it's deciding where to start.
The Practical Question
The headlines this week were about $40B deals and frontier models. But for most Singapore SMEs, the relevant question isn't "Should I be using GPT-5 or Claude Opus 4.7?"
It's:
"Where in my business is someone manually doing what software could do?"
That's the question the week's news should prompt — not because the technology suddenly changed, but because the cost of not doing anything just got measurably higher.
At The Empyrean, we work with Singapore SMEs to identify exactly that — the practical, repeatable tasks where AI delivers value without disruption. If you're not sure where to start, we're happy to take a look at your operations and tell you honestly what would make sense.