The Bubble That Knows It's a Bubble
17 days ago
- #investment-trends
- #historical-patterns
- #AI-bubble
- Sam Altman, OpenAI CEO, warns about AI investment bubble, causing market drops.
- MIT study shows 95% of companies investing in generative AI see no returns.
- AI investment consumes over half of America's capital expenditure.
- Examples of high valuations with little revenue: Anthropic ($4.1B), Character.AI ($1B), Inflection AI ($1.3B).
- Ray Dalio compares current AI hype to 1998-1999 dot-com bubble.
- Historical bubbles: Railway Mania (1840s), Radio stocks (1920s), Dot-com (1990s).
- Railway Mania saw 85% loss in share values but built UK's rail infrastructure.
- Dot-com bubble had NASDAQ P/E ratios of 200, leading to $5T in losses.
- AI bubble is global, with China, EU, and Saudi Arabia heavily investing.
- Pattern of bubbles: genuine breakthrough, early success, capital influx, social proof, reality check, infrastructure remains.
- Current AI bubble is more transparent, with real-time data and warnings.
- AI might differ due to rapid deployment, recursive improvement, and network effects.
- Bubble timeline suggests AI crash could occur in 4-6 years from 2022.
- Strategies for profiting post-bubble: focus on infrastructure, real revenue, necessity, and fire sales.
- Modern tools like regulation, risk management, and communication may mitigate bubble impacts.
- AI could disrupt jobs during its rise, not just after a crash.
- Key lesson: maintain patience and perspective during euphoria and despair.