The VC landscape has split into two extremes. A few elite firms and sovereign wealth funds are funding mega-rounds for about 20-30 top AI companies, while the broader ecosystem of seed funds, Series A specialists, and new managers is getting crushed by a lack of capital and liquidity.

Similar to the dot-com era, the current AI investment cycle is expected to produce a high number of company failures alongside a few generational winners that create more value than ever before in venture capital history.

The current AI boom isn't just another tech bubble; it's a "bubble with bigger variance." The potential for massive upswings is matched by the risk of equally significant downswings. Investors and founders must have an unusually high tolerance for risk and volatility to succeed.

Sam Lessin predicts massive losses for seed VCs backing companies branded as "AI businesses." These ventures are too capital-intensive and commoditizable to generate traditional venture returns, even if they become massive. AI should be a tool, not the business model itself.

The current AI investment cycle will likely result in a power-law distribution of returns. A small group of roughly 25 "magnificent" private companies will generate most of the value, while the vast majority of other AI startups will lead to significant capital deprecation for their investors.

Aggregate venture capital investment figures are misleading. The market is becoming bimodal: a handful of elite AI companies absorb a disproportionate share of capital, while the vast majority of other startups, including 900+ unicorns, face a tougher fundraising and exit environment.

The current AI investment climate feels as 'risk-free' as the 2021 bubble. Venture firms are likely using flawed loss-ratio models, underestimating how many AI 'unicorns' will fail to generate returns, just as they did with the B2B SaaS unicorns from the previous cycle.

The increased volatility and shorter defensibility windows in the AI era challenge traditional VC portfolio construction. The logical response to this heightened risk is greater diversification. This implies that early-stage funds may need to be larger to support more investments or write smaller checks into more companies.

The venture capital landscape is experiencing extreme concentration, with a handful of AI labs like OpenAI and Anthropic raising sums that rival half of the entire annual VC deployment. This capital sink into a few mega-private companies is a new phenomenon, unlike previous tech booms.