The New Power Law: Why AI is Rewriting Capital Allocation

The traditional venture capital power law is no longer a niche industry feature. It is a systemic reality of the modern economy. AI is not just an incremental software upgrade. It is a fundamental shift that allows capital to be converted directly into compute, which in turn compounds product quality and market dominance. This transition creates a winner-take-most dynamic that makes conventional diversification strategies obsolete. For institutional allocators and investors, the implication is clear: access to the top 1% of category-defining companies is no longer just a goal. It is a requirement for avoiding sub-market returns. Those who fail to adapt their portfolio construction to favor high-conviction, core-AI allocations will find themselves structurally disadvantaged, as the largest opportunities migrate toward physical infrastructure, energy, and labor-intensive sectors that legacy software models could never touch.

The End of Software as a Proxy for Value

For decades, venture capital operated on the assumption that software was the ultimate scalable asset. However, as David George and Aram Verdiyan note, AI is attacking the 30 trillion dollars of global GDP previously untouched by traditional tech, including labor, healthcare, and transportation. The hidden consequence here is that AI is not just a new category. It is an expansionary force.

AI is attacking every facet of the GDP, transportation, labor, services, capital coordination. There hasn't been a technology paradigm that hits on 30 trillion in GDP at the same time.

-- Aram Verdiyan

Most investors mistakenly view the rise of AI through the lens of zero-sum competition between labs and applications. Systems thinking reveals the opposite: the market is expanding so rapidly that both layers are growing concurrently. The immediate, visible benefit of AI is faster code generation, but the downstream effect is a complete reimagining of knowledge work. This creates a lasting advantage for those who stop treating AI as a satellite position and recognize it as a core allocation.

Why the Obvious Fix Makes Things Worse

Conventional wisdom suggests that startups should be lean to avoid the overhead of excessive capital. In the AI era, this logic is inverted. Because frontier AI companies can convert dollars directly into compute, and compute directly into superior product, capital has become a competitive moat rather than a liability.

The system responds to this dynamic by forcing a death of the middle. Firms that lack the scale to provide operational resources or the deep domain expertise to de-risk outcomes for founders are increasingly irrelevant. Founders are demonstrating a clear preference for partners who can help navigate the complexities of scaling, leaving mid-tier firms with diminished access to the category winners that drive the power law.

For the first time in my career you can take capital and throw it at a company and it compounds their advantage... You can throw dollars at compute and compute can make products and the business is better.

-- David George

The 18-Month Payoff Nobody Wants to Wait For

The most significant friction for allocators is the timeline to liquidity. Institutional investors often prioritize immediate returns, but the best-performing firms consistently hold category-defining companies through their compounding phases. The temptation to exit early, often driven by the desire to lock in gains, is a tactical error that sacrifices long-term enterprise value.

Furthermore, the AI-native transition is creating a bifurcation in private equity. Companies that simply infuse AI into legacy workflows without re-engineering their core systems are finding that they cannot compete with AI-native management teams. This creates a hidden cost: companies that attempt to bolt on AI without building from the studs risk churn and NPS degradation, which compounds into financial failure when coupled with debt.

Key Action Items

• Shift from Diversification to Concentration: Over the next 12 to 18 months, re-evaluate portfolios to ensure capital is concentrated in the top 1% of category-defining firms, rather than spread across 50+ holdings.
• Prioritize AI-Native Over AI-Infused: When evaluating growth-stage investments, distinguish between companies that are fundamentally re-architected for AI and those merely adding AI features. The latter face significant churn risks.
• Audit Supply-Side Bottlenecks: Look for investments in the left side of the AI stack, such as energy generation, grid infrastructure, and data center efficiency. This is where the next 100-billion-dollar opportunities reside.
• Embrace Longer Liquidity Horizons: Accept that the best AI companies will stay private longer to maximize compounding. Shift expectations from short-term distributions to long-term capital appreciation.
• Focus on Task-Level Utility: Move beyond ARR metrics in early-stage AI. Deeply analyze usage patterns and customer adoption texture to distinguish hype from real, resilient economic value.