The complex effects of AI are causing traditional market relationships, like yields reacting to economic surprises, to break down. In this new regime, broad diversification and passive strategies are ineffective as winners and losers become more distinct and dispersion explodes.
Related Insights
An LP's diversification strategy across different venture funds is undermined when every fund converges on a single theme like AI. This creates a highly correlated portfolio, concentrating systemic risk rather than spreading it. The traditional diversification benefits of investing across multiple managers, stages, and geographies are nullified.
Active managers are struggling against the S&P 500 not just from bad picks, but because the market is dominated by a few AI stocks they can't fully concentrate in. Many also became too defensive during April's volatility, causing them to miss the subsequent sharp market rebound.
Similar to the dot-com era, the current AI investment cycle is expected to produce a high number of company failures alongside a few generational winners that create more value than ever before in venture capital history.
While investors now believe in AI's transformative power, it remains unclear who will profit most. Value could accrue to chip makers (NVIDIA), foundation models (OpenAI), or the application layer. This fundamental uncertainty is a primary driver of the significant volatility across the tech sector.
For the first time, the high-multiple software industry faces a potential existential threat from AI. Even the possibility of disruption is enough to compress valuations, causing massive dispersion where indices look calm but underlying sectors are experiencing extreme rotation.
Today's market is more fragile than during the dot-com bubble because value is even more concentrated in a few tech giants. Ten companies now represent 40% of the S&P 500. This hyper-concentration means the failure of a single company or trend (like AI) doesn't just impact a sector; it threatens the entire global economy, removing all robustness from the system.
Initially, the market crowned OpenAI (via proxies Nvidia/Microsoft) the definitive AI leader. Now, with Google and Anthropic achieving comparable model performance, the market is re-evaluating. This volatility shows investors moving from a "one winner" thesis to a landscape where top AI models are becoming commoditized.
David Kaiser of Methodical Investments posits a contrarian view on AI's market impact. Instead of creating perfect efficiency, he argues AI and the data it processes might actually create more mispricings and inefficiencies. This provides opportunities for disciplined, rules-based strategies that don't constantly adapt to short-term noise.
The underperformance of active managers in the last decade wasn't just due to the rise of indexing. The historic run of a few mega-cap tech stocks created a market-cap-weighted index that was statistically almost impossible to beat without owning those specific names, leading to lower active share and alpha dispersion.
Drawing a parallel to the early internet, where initial market-anointed winners like Ask Jeeves failed, the current AI boom presents a similar risk. A more prudent strategy is to invest in companies across various sectors that are effectively adopting AI to enhance productivity, as this is where widespread, long-term value will be created.
