AI Concentration Threats in the S&P 500 and How to Diversify

AI is prompting a significant shift in investment strategy, even for those with decades of successful investing experience. The traditional reliance on the S&P 500 index fund, which has historically yielded an average of over 10% per year through passive investing, is being re-evaluated due to the increasing concentration of AI-driven companies within the index.

The S&P 500 and the AI Concentration Risk

The S&P 500 index fund, a popular passive investment vehicle, automatically allocates money across the top 500 US companies. While this strategy has historically been robust, protecting investors even if some stocks crash, a new concern has emerged: AI.

Currently, 40 cents of every dollar invested in the S&P 500 goes into just 10 companies: Nvidia, Microsoft, Apple, Alphabet, Amazon, Broadcom, Meta, Tesla, Berkshire Hathaway, and JP Morgan. Nvidia alone accounts for 7 to 8 cents of every dollar. This concentration is due to the S&P 500 being market-cap weighted, meaning more valuable companies take up a larger slice of the fund.

Excluding Berkshire Hathaway and JP Morgan, the remaining companies share a common trait: aggressive spending on AI. While AI is seen as the future, the current valuations of these AI-focused companies are based on projected future earnings rather than current revenue. For these companies to justify their valuations, they would need to generate approximately $2 trillion in revenue, which is more than Nvidia, Microsoft, Apple, Alphabet, Amazon, and Meta combined in 2024. This reliance on future potential, especially when linked to a single technology, raises concerns.

Many of these companies are making a significant bet on AI and financing it with substantial debt. For example, Sam Altman has committed $1.4 trillion to AI infrastructure, despite OpenAI's current annual revenue being only $13 to $20 billion. This debt is often hidden, with companies boosting valuations by circulating money among themselves, creating a self-reinforcing feedback loop. Larger companies attract more passive investment, driving up their prices and increasing their share of the S&P 500, which in turn provides more capital from passive investors.

This phenomenon also creates an illusion of economic strength. Deutsche Bank data suggests that without current AI spending, the US economy might already be in a recession.

Alternative S&P 500 Strategies: Equal-Weighted Funds

One potential solution to the concentration risk is an equal-weighted S&P 500 index fund. In such a fund, every company is treated equally, meaning the top 10 stocks would represent around 2% of the fund instead of 40%. This approach reduces exposure to a potential AI bubble.

However, equal-weighted funds have a drawback: they follow a "negative momentum" approach. When a company's stock performs well, the fund automatically sells some of it and uses the proceeds to buy companies that haven't performed as well, effectively selling winners and buying losers to maintain balance. In contrast, a market-cap weighted S&P 500 fund allows winning companies to take a larger share of the index without frequent trading.

Frequent trading in equal-weighted funds leads to higher costs, which can erode gains. Therefore, while reducing S&P 500 exposure, the majority of a stock market portfolio might still be best kept in the regular market-cap weighted version.

Diversifying Beyond the US Market

Historically, global economic dominance shifts, and no single country remains on top forever. While the US market has dominated for the past 14 years, this has not always been the case. For example, in 1900, the United Kingdom held 24% of the global stock market. Later, in the late 1980s and early 90s, Japan's economy was booming while the US struggled.

Investing solely in the S&P 500 means missing out on global companies like TSMC, Samsung, Toyota, Tencent, AstraZeneca, and HSBC, which are driving significant growth worldwide.

A global stock market fund, such as VWRP, offers exposure to approximately 3,700 to 3,800 companies across over 45 developed and emerging countries, including the US, UK, Europe, Japan, China, and India. While these funds may still be heavily weighted towards US companies initially, they automatically rebalance over time as global economic power shifts. These funds also typically have low expense ratios, such as 0.19% annually. For US-based investors, Fidelity's international fund (ticker: FSPSX) is a comparable option.

Platforms like Trading 212 offer "model pies," which are pre-built, globally diversified portfolios designed by professional asset managers, allowing users to choose a risk level.

Identifying Opportunities in the "Overlooked Zone"

The market can be categorized into four zones:

1. Crowded Zone: Large market cap, expensive valuations (e.g., top S&P 500 companies like Nvidia, Tesla, Meta). Everyone, including passive investors, is already invested here.
2. Defensive Zone: Companies with stable demand and predictable earnings (e.g., McDonald's, Walmart, Coca-Cola). Valuations are not low but offer good value due to consistent cash generation.
3. Speculative Zone: Smaller, highly hyped companies with valuations far exceeding current revenue (e.g., Beyond Meat, Peloton, AMC during their hype phases). These are considered high-risk, momentum-driven investments.
4. Overlooked Zone: This is where significant opportunities, especially with AI, are believed to exist.

Companies in the crowded zone are betting on developing the best AI model to capture all profits. However, a growing trend among startups and small-to-mid-cap companies is to offer users a choice between different AI models (e.g., Claude, ChatGPT, Gemini) within their software. As these models become increasingly similar in ability, price may become the deciding factor, leading to price wars that benefit users but not the large companies that invested billions in development.

The real value, when new technology becomes interchangeable, shifts from the developers to the companies that use it most intelligently. Therefore, investing in small and mid-cap businesses that apply AI to real problems at a lower cost, without carrying massive debt, presents a significant opportunity. This involves investing in small and mid-cap funds and actively seeking AI startups.

Increasing Gold Reserves

Technological leaps create both opportunities and instability, impacting power, currencies, and trust in the system. Consequently, increasing gold reserves is a strategic move. China, for instance, has been aggressively buying gold, with levels significantly increasing since 2023. Many believe China aims to challenge the US dollar's dominance by making gold usable through initiatives like the "gold corridor" for international trade, particularly with BRICS nations.

Globally, central banks have increased their gold holdings, with gold surpassing US Treasuries as the largest foreign reserve asset for the first time since 1996. This shift is partly driven by the reclassification of gold as a Basel III Tier 1 asset in 2025, allowing it to be treated similarly to cash or US Treasuries on a bank's balance sheet. This change has prompted financial institutions to increase their gold reserves from around 20% to a recommended 30%, driving up demand for a finite resource. Some predict gold's value could double in five years.

Investing in gold can be done through physical gold for long-term insurance or by dollar-cost averaging into an iShares Physical Gold ETF. The strategy is not to go "all in" but to allocate a small, strategic portion of the portfolio to gold, mirroring the actions of countries and major financial institutions.

Maintaining a Cash Reserve

A common mistake, especially among young investors, is not having a sufficient cash reserve. This can force them to sell investments at inopportune times. Even experienced investors like Warren Buffett are increasing their cash piles, with Berkshire Hathaway holding $347.7 billion in cash and cash equivalents as of Q1 2025. This reflects a value investor's approach of waiting for attractive opportunities.

Holding more cash provides a buffer against market crashes and allows for opportunistic buying when prices are low. Developing money-making skills, particularly during economic downturns, can also provide capital for investment.

Conclusion

The current investment strategy involves a diversified approach to navigate potential AI-driven market shifts:

• Continue investing in the S&P 500: To benefit from continued AI-driven growth.
• Diversify into global markets: To mitigate single-country risk and capture worldwide growth.
• Invest in small and mid-cap stocks: To capitalize on the "overlooked zone" opportunities in AI application.
• Increase gold reserves: As a hedge against systemic instability and currency shifts.
• Maintain a cash reserve: For financial security and opportunistic investing during market downturns.

This strategy acknowledges market uncertainty and prepares for various outcomes rather than attempting to predict them.