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 primary bear case for specialized neoclouds like CoreWeave isn't just competition from AWS or Google. A more fundamental risk is a breakthrough in GPU efficiency that commoditizes deployment, diminishing the value of the neoclouds' core competency in complex, optimized racking and setup.

Many NeoClouds are over-leveraged with loans based on volatile market caps and rely heavily on high-risk startups. This creates a fragile economic model akin to a mortgage crisis, where customer defaults could trigger a cascade of financial problems. Lightning AI mitigates this by being debt-free and focusing on enterprise clients.

The severe AI compute shortage has turned cloud providers into kingmakers. Instead of simply auctioning compute to the highest bidder, they are forced to make judgment calls on which AI startups ("Neo Labs") they believe in, effectively acting as venture capitalists by allocating the most critical resource for survival.

Historical tech cycles show that 95-99% of companies fail. For most current AI startups, the next 12-18 months represent a value-maximizing moment to sell before their technology is commoditized or outcompeted by foundation models.

The next market correction in AI won't be from a general oversupply of GPUs. Instead, it will stem from the fragmentation of smaller players building their own data centers. These niche clouds will struggle for customers, leading to a debt crisis and eventual reconsolidation back to a few major players.

Startups training foundation models face a new existential threat: the death of on-demand compute. Cloud providers, leveraging scarcity, now push for expensive three-to-five-year contracts. This forces early-stage companies into massive, long-term commitments they can ill afford and whose future needs are highly uncertain.

The dot-com era saw ~2,000 companies go public, but only a dozen survived meaningfully. The current AI wave will likely follow a similar pattern, with most companies failing or being acquired despite the hype. Founders should prepare for this reality by considering their exit strategy early.

The AI ecosystem has over 75 'NeoLabs' spun out from frontier research labs, and two-thirds are projected to be worth nothing. Early funding was based on talent alone, but the market has shifted to demand a viable business model and a clear path to revenue. For these companies, the "next round's a bitch."