Are You Holding the Right Assets?
Ali Mohammed
Chief Executive Officer @ Ramad Bridge LLC | Building Payment Corridors in East Africa
May 17, 2026
.
01 — The Signal Nobody Missed. Except Most People Did.
The S&P 500 returned +31% between May 2025 and May 2026. That's a solid year by any measure. But that's not the story.
The story is what happened in the layer below the index. Chip companies. Data center infrastructure. Clean energy feeding AI compute. That layer didn't return 31%. It returned +320% on average for chips. +419% on average for data center and energy names. Some individual names: Bloom Energy +1,647%. Micron +770%. Intel +483%.
If you were positioned in what I call the Singularity Loop — chips plus infrastructure plus energy — you didn't just beat the market. You lapped it. Repeatedly.
So. What exactly is the Singularity Economy? It's not a buzzword. It's a structural shift in how value gets created. The moment AI systems start compounding — getting faster, cheaper, and smarter on their own — the infrastructure that runs them becomes the most valuable real estate on the planet.
02 — How the Loop Works. And Why It Keeps Paying.
Think of it like a telecom rollout, but faster and more concentrated. When mobile networks scaled in the 2000s, three layers got paid: tower infrastructure, device hardware, and the software on top. The Singularity Loop is the same architecture, just compressed into a decade and running at a pace nobody expected.
Here's the loop in plain language:
- More capable models — AI models get more capable. That's step one. It keeps happening.
- Need more chips. — More parameters. Bigger training runs. More inference calls per day. Chips are the physical substrate of intelligence. No chips, no AI.
- Chips need data centers. — You can't run hyperscale chip clusters in your office. Purpose-built data centers — cooled, powered, connected — are the housing layer.
- Data centers need power. — A single AI training cluster can consume as much electricity as a small city. New power sources — nuclear, clean gas, grid-scale storage — are not optional. They are load-bearing.
- So when AI demand grows, — Every link in the chain got more valuable. That's why the numbers look the way they do.
Global corporate AI investment hit $581.7 billion in 2025 — up 130% year over year (Stanford HAI, 2026 AI Index Report). Q1 2026 broke every venture capital record ever set. $300 billion deployed in a single quarter. Of that, $242 billion — 80% — went to AI companies. This is not speculative capital. This is infrastructure-grade conviction.
03 — The Gap Between the Old Economy and the New One.
Real estate returned +5%. Financials returned +11%. Health care +9%. These are not bad numbers. They're just not the same category of outcome as what happened in the Singularity Loop.
Technology — the broadest sector — returned +76%. That sounds impressive until you remember that the Singularity Loop averaged +320% to +419% over the same period. The gap between the best traditional sector and the Singularity Loop wasn't a gap. It was a wall.
This isn't a criticism of traditional assets. It's a signal about capital allocation. Every major economy has a finite amount of investable attention. Right now, a large and growing portion of that attention is pointing at one thing: the infrastructure that runs AI. The chart above shows you where everything else was standing while that happened.
The gap will likely keep widening. AI spending is projected to reach $2.52 trillion in 2026 alone (Gartner). Workers with AI skills already earn 56% more than peers without them (PwC). By 2030, AI is expected to add $15.7 trillion to global GDP. The numbers aren't speculative forecasts anymore. They're forward-looking signals from capital already in motion.
04 — The Labs Building the Engine. What They're Worth Now.
The AI labs at the frontier — the companies training the largest, most capable models — are not publicly traded. Most people can't buy them directly. But their valuations tell you something important about how the market prices the future.
In one year:
- OpenAI: from $300B to $852B. That's +184%.
- Anthropic: from $61.5B to approximately $1 trillion. That's +1,526% in 12 months.
- xAI: from $80B to $250B. That's +213%.
- Mistral AI: from $6.2B to $14B. That's +126%.
These valuations are private estimates based on funding rounds — not audited figures, not market prices. But they reflect real capital from informed investors making large, concentrated bets on specific outcomes. Anthropic's Q1 2026 round alone was $30 billion. OpenAI's was $122 billion — the largest private funding round in history.
You can't buy Anthropic stock. But you can understand what these valuations signal about the demand for chips, compute, and power. Every trillion-dollar AI lab needs a multi-trillion-dollar supply chain beneath it. That supply chain is publicly traded. That's the Singularity Loop.
05 — Africa Is Not Outside This Story. You're Not Outside This Story.
Here's something most people don't know. The top 10 countries by everyday AI adoption rate per KPMG? All emerging markets. #Nigeria is on that list. #India. #Egypt. #Brazil. The tools being built by trillion-dollar labs in San Francisco are being tested and used daily by people in #Lagos, #Nairobi, #Cairo, #Johnstown, #Mogadishu and #Accra.
The infrastructure value in the Singularity Economy — chips, compute, energy, models — is being accumulated right now by people who understand the loop. The people who built and held telecom infrastructure in the 1990s didn't look smart until the mobile wave hit. The people who understood data centers in 2010 didn't look smart until cloud computing took over. The pattern is repeating.
This isn't about whether you should go buy Nvidia tomorrow. That's between you and a qualified financial advisor. This is about situation awareness. Do you have a clear picture of where generational wealth is being accumulated in the current economy? If not, that gap in awareness is a risk in itself.
The Singularity Economy is already running. The loop is already paying. Africa's engineers, founders, and investors are already using the tools. The only question left is whether you're holding any part of the infrastructure that makes all of it work.