AI Power Law: Venture Capital Strategy Shifts
An analysis of how AI is fundamentally altering venture capital power laws, making capital a direct driver of competitive advantage. The discussion covers the necessity of core asset allocation, the death of the middle in VC firms, and the critical importance of access and sizing for institutional investors.
The Structural Shift in AI Value Creation
The venture capital landscape is undergoing a fundamental transformation driven by the unique economics of Artificial Intelligence. Unlike previous technology cycles, AI firms exhibit a direct correlation between capital deployment and competitive advantage. As noted by industry leaders, capital can now be thrown at a company to purchase compute, which directly improves product quality and reinforces the moat of frontier players. This dynamic has made the power law more extreme than at any point in the last two decades, with only 20 out of 3,000 U.S. venture firms achieving consistent 3x net returns over the past 20 years.
Implications for Institutional Allocation
For institutional investors, this shift necessitates a re-evaluation of asset allocation strategies. AI is no longer a satellite position but a core allocation, as it impacts every facet of the GDP, including transportation, labor, and services. The addressable market for AI is estimated to be 10x larger than traditional software because it targets the economic value of labor tasks rather than just software licenses. Allocators must recognize that missing access to the top 5-10 frontier companies results in significantly underperforming returns compared to the average venture outcome of 1-2x net.
Portfolio Construction and Sizing
Effective portfolio construction in this environment requires a focus on access, selection, and sizing. The "death of the middle" is evident as mid-sized VC firms struggle to compete with large, resource-rich partners who can de-risk outcomes for founders. To succeed, late-stage investors must size positions at 5-10% of their fund in category-defining companies to achieve fund-returning liquidity. This strategy is only viable for firms with strong early-stage franchises that provide the necessary access and relationship depth. Conversely, smaller firms must specialize in pre-seed or niche verticals to avoid direct competition with giants.
Future Outlook and Bottlenecks
The next wave of value creation will likely emerge from categories that barely exist today, such as robotics, autonomy, and healthcare. However, the primary bottleneck for AI growth is no longer demand but supply, specifically energy, data centers, and chips. Investors are advised to look beyond software to these physical infrastructure plays, where the U.S. faces significant regulatory and transmission challenges. The consensus is that the current AI cycle is not a bubble but a structural shift, with the potential for market caps exceeding $100 trillion in the coming decade.
Key insights
-
AI firms convert capital directly into compute, which improves product quality and compounds competitive advantages, unlike traditional startups where excess capital creates coordination issues.
Impact: This dynamic justifies larger capital deployments in AI and explains the extreme power law observed in current venture returns.
-
Only 20 out of 3,000 U.S. venture capital firms have achieved consistent 3x net returns over the last two decades, indicating that access to category-defining companies is the primary driver of success.
Impact: Institutional investors must concentrate their allocations in a small number of top-tier firms to avoid averaging out to sub-par returns.
-
The addressable market for AI is 10x larger than traditional SaaS because it targets the economic value of labor tasks, not just software licenses.
Impact: This expands the potential for high-growth companies and justifies higher valuations for AI-native businesses.
-
Late-stage venture investing now requires sizing positions at 5-10% of the fund in a single category winner to achieve fund-returning liquidity, a strategy enabled by early-stage relationships.
Impact: This shifts the value proposition of late-stage funds from diversification to concentrated exposure to high-conviction winners.
-
The primary bottleneck for AI growth is supply-side constraints, including energy, data centers, and chips, rather than demand.
Impact: Investment opportunities are shifting toward physical infrastructure and energy solutions, which are critical for scaling AI capabilities.
Action items
-
Reclassify AI from a satellite to a core asset allocation in institutional portfolios, reflecting its impact on every facet of GDP.
Impact: This ensures adequate exposure to the highest-growth sector and avoids underperformance relative to the market.
-
Concentrate venture allocations in the top 20-50 firms with a proven track record of accessing category-defining companies.
Impact: This mitigates the risk of averaging out to sub-par returns and maximizes exposure to power-law outcomes.
-
For late-stage investors, size positions at 5-10% of the fund in a single category winner to achieve fund-returning liquidity.
Impact: This strategy leverages the extreme power law to generate significant returns from a small number of high-conviction bets.
-
Evaluate AI-native companies based on their ability to replace labor tasks, not just software licenses, to accurately assess their TAM.
Impact: This provides a more accurate valuation framework for AI companies and identifies those with the largest potential upside.
-
Invest in supply-side infrastructure, including energy, data centers, and chips, to address the primary bottlenecks for AI growth.
Impact: This positions investors to benefit from the physical constraints that will determine the pace of AI adoption and scaling.
Quotes
“For the first time, you can take capital and throw it at a company and it compounds their advantage.”
“We've looked at the data of 3,000 venture capital firms in the U.S. Only 20 have achieved consistent 3x net returns over the last two decades.”
“AI is attacking every facet of the GDP. Transportation, labor, services, capital, coordination.”