From Dry Powder to AI Power: PE’s Technology Infrastructure Play

Private equity firms are sitting on $2.5 trillion in dry powder. That’s a staggering number that gets thrown around in every industry presentation.

But here’s what’s more interesting: where that capital is actually going.

Advent just acquired LayerZero Power Systems to capitalize on AI data center growth. Apollo committed $15 billion to data center infrastructure with OpenAI. KKR’s infrastructure funds are pivoting toward digital infrastructure.

The pattern is clear: PE firms are betting big on the picks and shovels of the AI revolution. And for good reason.

While everyone else is trying to pick the next AI unicorn, the smartest money is investing in the infrastructure that makes AI possible.

1. The Infrastructure Behind the AI Boom

Let’s start with a simple truth: artificial intelligence doesn’t run on magic. It runs on infrastructure.

Massive infrastructure.

The AI infrastructure stack includes:

• Power systems: AI data centers consume 10-50x more power than traditional data centers
• Cooling systems: Advanced cooling technology to handle extreme heat loads
• Network infrastructure: Ultra-low latency connections and massive bandwidth
• Physical security: Protecting billions of dollars in computing equipment
• Backup systems: Redundancy and disaster recovery for critical AI workloads

Each layer of this stack represents a massive investment opportunity. And unlike AI software companies—where valuations are often based on hope and hype—infrastructure investments are backed by long-term contracts and predictable cash flows.

Consider the numbers: a single AI data center can cost $10+ billion to build and requires ongoing infrastructure investment throughout its lifecycle. Multiply that by the hundreds of AI data centers being planned globally, and you’re looking at a trillion-dollar infrastructure buildout.

2. Why Data Centers Are the New Real Estate

For decades, private equity real estate focused on office buildings, retail centers, and industrial warehouses. Those asset classes aren’t going away, but they’re not where the growth is.

The growth is in data centers. Specifically, AI-optimized data centers.

What makes AI data centers different:

• Power density: 10-50x higher power consumption per square foot
• Cooling requirements: Sophisticated cooling systems that can cost more than the building itself
• Network connectivity: Direct fiber connections to major cloud providers and AI companies
• Security: Physical and digital security requirements that exceed traditional data centers
• Scalability: Ability to rapidly expand capacity as AI workloads grow

But here’s the key insight: AI data centers aren’t just real estate investments. They’re technology infrastructure investments with real estate characteristics.

The best AI data center investments combine:

• Prime locations: Proximity to power generation and network infrastructure
• Technology expertise: Understanding of AI workload requirements and cooling technology
• Utility relationships: Partnerships with power companies to secure long-term capacity
• Customer relationships: Direct relationships with AI companies and cloud providers

3. LayerZero and the Critical Power Play

Advent’s acquisition of LayerZero Power Systems is a perfect example of how to think about AI infrastructure investing.

LayerZero doesn’t build data centers. They build the critical power infrastructure that makes AI data centers possible. Static transfer switches, power distribution units, and backup power systems.

Why this matters:

• Mission-critical: AI workloads can’t tolerate power interruptions
• High barriers to entry: Specialized technology and regulatory requirements
• Recurring revenue: Ongoing maintenance and upgrade cycles
• Global opportunity: AI infrastructure buildout is happening worldwide

The LayerZero deal illustrates a key principle: sometimes the best way to invest in a technology revolution is to invest in the infrastructure that enables it, not the technology itself.

Other critical infrastructure plays include:

• Cooling technology: Advanced cooling systems for high-density computing
• Network equipment: Ultra-low latency networking for AI workloads
• Security systems: Physical and digital security for AI infrastructure
• Monitoring and management: Software and services to manage complex AI infrastructure

4. The Risk: When Infrastructure Becomes Commodity

But let’s be honest about the risks.

Not every infrastructure investment will benefit from the AI boom. And even good infrastructure investments can become commoditized if you’re not careful.

The commoditization risks:

• Technology obsolescence: AI infrastructure requirements are evolving rapidly
• Oversupply: Too much capital chasing the same infrastructure opportunities
• Standardization: As AI infrastructure matures, differentiation becomes harder
• Regulatory changes: Environmental and zoning regulations could impact data center development

The key to avoiding commoditization is to focus on infrastructure investments with sustainable competitive advantages:

• Proprietary technology: Unique capabilities that are hard to replicate
• Strategic locations: Prime real estate that can’t be easily duplicated
• Customer relationships: Long-term contracts with major AI companies
• Operational expertise: Deep understanding of AI infrastructure requirements

5. Positioning Your Firm for the Next Wave

So how do you position your firm to benefit from the AI infrastructure boom?

Start with the value chain analysis. Map out the entire AI infrastructure value chain and identify where your firm has competitive advantages. Don’t just follow the crowd into data center real estate.

Develop technical expertise. AI infrastructure is complex and evolving rapidly. Invest in building technical expertise within your team or through advisory relationships.

Think globally. The AI infrastructure buildout is happening worldwide, but regulatory and market conditions vary significantly by region. Look for opportunities in markets that others might overlook.

Focus on mission-critical components. The best AI infrastructure investments are in components that are absolutely essential and hard to substitute. Power systems, cooling technology, and network infrastructure are good examples.

Build ecosystem relationships. The AI infrastructure ecosystem is highly interconnected. Relationships with AI companies, cloud providers, and technology vendors can provide deal flow and value creation opportunities.

Consider the full lifecycle. AI infrastructure isn’t just about initial buildout. There are ongoing opportunities in maintenance, upgrades, and eventual decommissioning.

The Bottom Line

The AI revolution is creating a massive infrastructure investment opportunity. But like any investment theme, success depends on execution, not just timing.

The firms that succeed in AI infrastructure will be those that:

• Understand the technical requirements and competitive dynamics
• Focus on sustainable competitive advantages, not just market growth
• Build deep relationships within the AI ecosystem
• Think beyond just data center real estate to the entire infrastructure value chain

The firms that just chase the AI infrastructure theme without understanding the underlying dynamics will likely be disappointed.

But for firms that do their homework and invest thoughtfully, the AI infrastructure boom represents one of the largest infrastructure investment opportunities of our generation.

How is your firm thinking about AI infrastructure investments? What opportunities are you seeing beyond traditional data center real estate? Share your insights with the VCII community.