Stories
Dot-Com Bubble vs AI Boom: Are Markets Heading for a Repeat of 2000?
- •Market concentration in the S&P 500's top ten AI-linked companies has surpassed the peak levels seen during the dot-com bubble in 2000, reaching roughly 38 to 41 percent.
- •Valuation metrics like the Buffett Indicator, which stands at over 230 percent, and the Shiller CAPE ratio at around 41, show that today's market is among the most expensive in modern history.
- •Industry-wide spending on AI infrastructure heavily outweighs current revenues, with projected 2026 investments reaching about $1.5 trillion against a fraction of that in actual lab revenues.
- •Red flags such as circular financing, excess capacity building by tech giants like Meta, and the rapid rise of competitive Chinese AI labs pose growing threats to Western valuations.
- •Experts advise investors not to blindly chase hype, but instead to focus on fair valuations, portfolio rebalancing, and global diversification across various asset classes.
Key Questions Answered
The top ten S&P 500 companies account for 38-41 percent of index value, surpassing the 26-27 percent concentration seen in 2000. Furthermore, the S&P 500 price-to-sales ratio is roughly 3.3 to 3.8 times compared to 2.3 times in 2000, and the Buffett Indicator stands near 230-237 percent versus 150 percent during the dot-com peak.
The AI industry's 2026 infrastructure spending is estimated at roughly $1.5 trillion, whereas combined annualised revenues of top-funded labs like Anthropic and OpenAI total approximately $105 billion. OpenAI is also targeting up to $856 billion in cumulative compute spend through 2030, keeping expenditure significantly ahead of revenue targets.
Key risk signals include circular financing where top AI infrastructure firms invest in or guarantee each other's debt, such as NVIDIA reducing its initial $100 billion commitment to OpenAI to about $30 billion. Additionally, major capacity builders like Meta are building ahead of usage by offering to rent out surplus computing power, and valuation methods meant for steady businesses are being applied to highly unpredictable markets.
