The Technology Convergence Investment Framework

The Convergence Thesis#

Over the past decade, three technology platforms have evolved along mostly parallel trajectories: artificial intelligence (algorithmic capability), cloud infrastructure (compute and data), and enterprise software (workflow and organizational application). What I believe is underappreciated by most investors is that we are witnessing a convergence of these three domains that will reshape how companies operate and compete.

The companies that successfully navigate this convergence — those that are simultaneously excellent at AI, understand cloud infrastructure economics, and can embed themselves into enterprise workflows — will generate exceptional returns. Conversely, companies that operate primarily in one domain without bridging to the others will face margin pressure and competitive displacement.

---

Mapping the Convergence#

The Three Domains#

AI/Machine Learning: The algorithmic layer. This includes both the foundational models (large language models, computer vision models, etc.) and the fine-tuning, deployment, and inference infrastructure required to operationalize these models in production environments.

Cloud Infrastructure: The hardware and platform layer. This is where compute gets deployed, where data gets stored and processed, and where the economic models increasingly shift from capex to opex, from owned infrastructure to rented resources, from regional data centers to distributed global networks.

Enterprise Software: The application layer. This is where AI and cloud infrastructure become valuable to enterprises. A cloud platform with no compelling software applications is a utility; AI without integration into actual business processes is an academic exercise.

Where the Value Accumulates#

Historically, value accrual in technology has been concentrated in one layer:

• AI/ML research was primarily captured by labs and academic institutions
• Cloud infrastructure value was captured by AWS, Azure, Google Cloud
• Enterprise software value was captured by legacy monoliths (Oracle, SAP, Salesforce)

The convergence creates a fundamental shift: companies operating across multiple layers will capture disproportionate value.

Consider a simple example: A generic cloud provider offers compute capacity. A cloud provider that has embedded AI optimization (auto-scaling trained models, dynamic resource allocation based on workload patterns) captures incremental value. A cloud provider that then builds enterprise applications that leverage both cloud infrastructure AND AI generates even more value because it controls the entire stack and can optimize end-to-end economics.

---

Investment Framework: Evaluating Convergence Plays#

When evaluating technology companies in this convergence environment, I use three core dimensions:

Dimension 1: Depth (How Deep Into Each Layer Does the Company Operate?)#

Companies scoring 8+ on this framework have the potential for exceptional value creation because they can optimize across layers. Microsoft, for example, can embed AI models into Azure infrastructure and then surface them through Office applications — creating a seamless enterprise experience.

Dimension 2: Breadth (How Many Enterprise Use Cases Can the Company Serve?)#

Even companies strong across all three layers can underperform if they're focused on narrow use cases. Conversely, companies that can apply their AI + cloud + software stack to multiple industries and workflows will capture more value.

Microsoft's strength here is evident: the same underlying AI + cloud infrastructure can be applied to:

• Knowledge worker productivity (Copilot in Office)
• Business intelligence (Power BI)
• Customer relationship management (Dynamics)
• Healthcare (Nuance acquisition)
• Financial services (compliance and fraud detection)

Companies with breadth can amortize their AI and infrastructure investments across larger TAM opportunities, driving superior unit economics.

Dimension 3: Durability (How Defensible Is the Moat?)#

Companies bridging multiple layers create compounding defensibility:

Cloud Infrastructure Lock-in: Once a customer deploys on Azure or AWS, switching costs are substantial. Add AI capabilities optimized for that platform, and switching becomes even more difficult.

Data Advantage: Companies operating at the application layer accumulate customer data that becomes more valuable when combined with AI and cloud infrastructure. This data advantage then feeds back into better AI models, further entrenching the platform.

Ecosystem Dynamics: Companies that open their platforms to third-party developers (as Microsoft has done with Copilot plugins, Azure Marketplace) create network effects that strengthen their position.

---

Identifying Convergence Winners#

Based on this framework, several characteristics indicate likely convergence winners:

1. Vertical Integration Signals: Companies acquiring or building capabilities in different layers signal management conviction in the convergence thesis. Microsoft's acquisition of Nuance (healthcare domain expertise) combined with Azure Health Data Services demonstrates this.

2. Developer Ecosystem Expansion: Companies expanding developer tools and SDKs across AI + cloud + software indicate they're preparing for convergence. AWS's acquisition of CodeGuru (AI-powered code review) is an example.

3. Margin Expansion Through AI: Companies showing gross margin expansion through AI-driven automation (like Azure with AI-optimized workloads) are capturing convergence value.

4. Cross-sell Acceleration: When companies can demonstrably drive upsell within their own ecosystem (Azure cloud adoption driven by Office integration, for example), it's a signal that convergence is creating value.

---

Practical Portfolio Application#

Overweight: Microsoft, Google (with Azure and AI gaps, slightly lower conviction), and software companies successfully embedding AI into their core platforms (Salesforce, Adobe, Workday).

Neutral/Selective: Cloud-only providers (Amazon, though AWS remains excellent) because they lack enterprise software distribution; AI-only companies (OpenAI, Anthropic) because they lack infrastructure and enterprise software integration.

Underweight: Legacy enterprise software companies that aren't successfully integrating AI and cloud infrastructure (Oracle remains challenged despite efforts); pure-play infrastructure companies without software application layer strategy.

---

The Five-Year Outlook#

By 2031, I expect the technology landscape will be characterized by 2-3 dominant convergence players (Microsoft being the clear leader, likely Google in second position) and a fragmented set of specialized companies operating in single layers or narrow convergence plays.

This consolidation will be positive for specialized companies that serve narrow use cases exceptionally well but negative for companies that sit in the middle of the convergence and can't differentiate on any dimension.

For investors, the framework suggests that:

1. Conviction on convergence leaders is warranted — these businesses will compound value across multiple layers
2. Understand where each company sits on convergence — don't assume growth will persist for single-layer players
3. Watch for convergence moves — M&A activity, API expansion, and new product launches can signal emerging convergence strategies

The companies that execute this convergence most effectively will be the generational winners of the next decade.