The AI bubble resembles the telecom bubble of the late 90s, where massive, real CapEx on physical infrastructure (fiber optic cables then, GPUs now) created real profits for suppliers. The danger is this euphoria, funded by cheap capital, leads to overinvestment with no guarantee of long-term profitability.

History shows that transformative technologies like railroads and the internet often create market bubbles. Investors can lose tremendous amounts of capital on overpriced assets, even while the technology itself fundamentally rewires the economy and creates massive societal value. The two outcomes are not mutually exclusive.

The frenzy in AI investment mirrors past technological revolutions like railways. Following Schumpeter's theory, overinvestment occurs as many firms race for dominance. This leads to a bust where most fail, but the infrastructure they built remains, benefiting society in the long run.

While many early investors in tech booms (e.g., telecom, AI) lose money, these 'bubbles' are not a societal waste. They fund the rapid construction of foundational infrastructure, like fiber optic networks or data centers, creating immense long-term value and options for future innovation that society ultimately benefits from.

The current massive capital expenditure on AI infrastructure, like data centers, mirrors the railroad boom. These are poor long-term investments with low returns. When investors realize this, it will trigger a market crash on the scale of 1929, after which the real value-creating companies will emerge.

The massive, redundant CapEx in AI infrastructure is analogous to the late-90s fiber-optic boom. While that fiber enabled future giants like Netflix, the initial investors went bankrupt. This suggests the ultimate beneficiaries of AI may be society and end-users, not the companies spending trillions on the build-out.

The current AI boom may not be a "quantity" bubble, as the need for data centers is real. However, it's likely a "price" bubble with unrealistic valuations. Similar to the dot-com bust, early investors may unwittingly subsidize the long-term technology shift, facing poor returns despite the infrastructure's ultimate utility and value.

Reid Hoffman argues the AI boom is not a bubble destined to collapse. The massive investment is creating valuable compute infrastructure with real demand. While specific company valuations may correct, it won't trigger the systemic contagion and debt crises associated with historical bubbles.

While disastrous for many investors, historical bubbles like the dot-com boom and railway mania left behind massively overbuilt infrastructure (fiber optics, rail networks). This infrastructure became cheap and abundant post-crash, enabling subsequent waves of innovation that benefited society for decades.