Nasdaq Composite: AI Experts Raise Red Flags as Valuations Stretch Higher

AI Bubble Concerns Mount as Valuation Warnings Escalate

Warnings of a speculative bubble in artificial intelligence (AI) equities are intensifying, with top financial voices drawing parallels to the late-1990s dot-com boom.

Between January and August 2025, figures including OpenAI CEO Sam Altman, Alibaba co-founder Joe Tsai, Bridgewater’s Ray Dalio, and Apollo Global’s Torsten Slok have raised red flags over unsustainable valuations.

With multi-billion-dollar AI infrastructure bets and revenue multiples far exceeding historical norms, traders are increasingly preparing for the risk of a sharp correction.

High-Profile Warnings Highlight Bubble Risk

Sam Altman delivered a blunt assessment on August 18, warning that “investors are overexcited about AI,” comparing current sentiment to classic bubble behavior. He cautioned that smart investors often “get overexcited about a kernel of truth,” especially as AI infrastructure spending approaches the trillion-dollar mark.

Joe Tsai echoed similar concerns at the HSBC Global Investment Summit in March, criticizing large-scale data center buildouts “on spec.” He flagged the $500 billion Stargate project as a sign that capital is outpacing real demand.

In January, Ray Dalio likened AI market sentiment to the 1998–99 Nasdaq rally, warning that inflated prices combined with rising interest rates could produce a similar outcome.

Apollo’s Torsten Slok followed with data showing that the S&P 500’s top 10 companies are now more overvalued than during the dot-com era, based on forward P/E ratios.

Valuation and Technical Indicators Confirm Excess

Market data reinforces these concerns. AI-related venture funding topped $100 billion in 2024, accounting for one-third of global VC flows. Palantir trades at a trailing P/E between 588x and 805x, while OpenAI’s $157–300 billion valuation reflects a 31–40x revenue multiple, despite continued losses.

Technical readings also point to overheated conditions. The S&P 500 Information Technology sector trades at a P/E of 38.4 — roughly 80% above long-term averages. Several AI stocks have RSI readings above 70, and the Nasdaq has gained 69% since ChatGPT’s launch, significantly outpacing the S&P 500’s 48% rise.

Historical Parallels Support Correction Risk

The setup mirrors the 1995–2001 dot-com rally, when the Nasdaq rose 400% before declining nearly 80%. Today’s AI firms, including OpenAI, are generating revenue — with projections above $12 billion for 2025 — but valuations remain stretched.

Capital is heavily concentrated in a small group of high-profile names, increasing exposure to downside pressure. A correction would likely unfold over a shorter time frame than the 15-year dot-com recovery, but the risk level remains elevated.

Short-Term Outlook: Cautious Bias with Measured Pullback Levels

While AI’s long-term potential is not in question, the Nasdaq’s current position suggests a measured short-term pullback is increasingly likely. At 21,623, the index is trading well above its key moving averages, which often act as natural reversion points during market cooldowns.

Traders should monitor the following near-term levels:

• 50-day moving average at 20,544 – a 5% decline from current levels.
• 20,000 area, including the June breakout zone – a 7–8% decline, where short-term support could emerge.
• 200-day moving average at 19,158 – would represent an 11–12% retracement, consistent with prior pullbacks during earnings season or position unwinding.

The trend remains intact above the 50-day moving average, but valuation premiums and technical signals indicate the potential for pressure on high-growth AI stocks. A 5–10% retracement would remain within historical norms and does not require a full reset of the current uptrend.

More Information in our Economic Calendar.