Core Conclusion

The current AI investment boom has structural similarities to the 2000 dot-com bubble, but with key differences. A simple “it will crash” analogy is insufficient, but valuations leave no margin for error.

Similarities to 2000

  • Capex growth far outpacing revenue growth
  • “This time is different” narrative dominates
  • Infrastructure investment leads, application monetization lags
  • Valuations heavily dependent on far-future expectations

Key Differences from 2000

  • Leading companies already have real earnings (NVDA, MSFT, GOOG)
  • AI technology has clear enterprise use cases
  • Leading companies are cash-flow rich, not pure cash-burning models

Risk Factors

  • AI capex return cycle uncertain
  • Open-source models may erode closed-source pricing power
  • Client-side AI adoption may be slower than expected