While investors now believe in AI's transformative power, it remains unclear who will profit most. Value could accrue to chip makers (NVIDIA), foundation models (OpenAI), or the application layer. This fundamental uncertainty is a primary driver of the significant volatility across the tech sector.

In the current AI boom, companies are raising subsequent funding rounds at the same high revenue multiples as previous ones, months apart. This is because growth rates aren't decelerating as expected, challenging the wisdom that valuation multiples must compress as revenue scales.

Major AI labs plan and purchase GPUs on multi-year timelines. This means NVIDIA's current stellar earnings reports reflect long-term capital commitments, not necessarily current consumer usage, potentially masking a slowdown in services like ChatGPT.

Despite bubble fears, Nvidia’s record earnings signal a virtuous cycle. The real long-term growth is not just from model training but from the coming explosion in inference demand required for AI agents, robotics, and multimodal AI integrated into every device and application.

The global economy's dependence on AI has created a massive concentration of risk in NVIDIA. Its valuation, exceeding the entire German stock market, makes it a single point of failure. A significant drop in its stock—which could still leave it overvalued—would have catastrophic ripple effects with nowhere for capital to hide.

While Nvidia dominates the AI training chip market, this only represents about 1% of the total compute workload. The other 99% is inference. Nvidia's risk is that competitors and customers' in-house chips will create cheaper, more efficient inference solutions, bifurcating the market and eroding its monopoly.

Despite massive investment in chips (NVIDIA) and models (OpenAI), it is not yet clear where long-term value will concentrate. The entire stack is in flux. Models could be commoditized by open source, chips could face historical commoditization cycles, and new AI-native apps could capture the most value. We are only in the early innings of a 30-year shift.

Contrary to common belief, the earliest AI startups often command higher relative valuations than established growth-stage AI companies, whose revenue multiples are becoming more rational and comparable to public market comps.

Swisher draws a direct parallel between NVIDIA and Cisco. While NVIDIA is profitable selling AI chips, its customers are not. She predicts major tech players will develop their own chips, eroding NVIDIA's unsustainable valuation, just as the market for routers consolidated and crashed Cisco's stock.