Valuing $NVIDIA as a Real Estate Company That Sells Housing to AI Agents ($100k/Share in 2034)

Determining the true worth of Nvidia stocks has been a journey of blending qualitative insights with hard numbers. I’ve taken on the task of performing a traditional discounted cash flow analysis, examining not just one, but two different scenarios: the pessimistic and the optimistic.

It’s crucial first to set the foundation with some qualitative data to contextualize these scenarios.

My initial valuation places Nvidia at $1,742 per share, which suggests that it’s significantly undervalued—by about 50%.

Investors could potentially see a compound annual growth rate of around 10% in the long term, with an even higher 30% in the most favorable conditions.

To understand what ‘pessimistic’ and ‘optimistic’ mean concretely, we’ll dive into Nvidia’s history, specifically the last decade of stock performance that has seen an exceptional appreciation, and explore whether such growth is sustainable.

The stock’s past performance has been stellar, reaching annual compound growth rates that rival the fastest-growing assets globally. But can this momentum be maintained? That’s what I aim to clarify by delineating three possible future scenarios—pessimistic, realistic, and optimistic.

With these scenarios, we will delve into a fair valuation for Nvidia using the discounted cash flow method, assigning weighted probabilities to each case to project an expected value.

I believe the most optimistic outlook warrants extra attention, as it could be pivotal for those who aim to be ahead of the curve in technological change.

In answering questions about the revenue and total addressable market for AI training, I’ve also scrutinized the relationship between the labor share of GDP and AI’s potential market reach—critical considerations for Nvidia’s valuation.

Valuation Overview – Combining Three DCF Models with Weighted Probabilities

	probability	fair value	weighted
failure	50%	$0	$0
pessimistic	20%	$758	$152
realistic	20%	$2,895	$579
optimistic	10%	$10,114	$1,011
			$1,742

Exploring the TAM for AI Training to Estimate Revenue Growth of Our Three Scenarios

In my analysis, I’ve assessed the value of Nvidia’s stock and concluded that it stands at $1,742 per share. This finding suggests that Nvidia’s shares are currently priced at half their actual value, according to the valuation method I’ve used.

Investors thinking long-term could expect a compound annual growth rate (CAGR) of around 10% annually. However, for those with a more optimistic outlook, the annual growth could potentially surge to about 30%.

Looking back, Nvidia’s historical stock performance over the past decade has been stellar, outpacing many other assets, even potentially surpassing Bitcoin.

It’s important, though, to set realistic expectations; as impressive as the growth has been (approximately 69% CAGR), it’s not sustainable at that level going forward.

To get a clearer picture, I’ve come up with three different scenarios—pessimistic, realistic, and optimistic—and assigned each a probability to calculate a weighted average value.

The most positive forecast stands out due to its impact on the assessment, and it warrants attention if we aim to stay ahead of the curve in technological changes.

When checking Nvidia’s potential revenue from the total addressable market for AI training, it’s essential to recognize AI’s role in the labor market.

AI is set to carve out a portion of the $51 trillion paid in wages worldwide. The AI market is divided into two segments: AI training, comparable to human education, and AI inference, similar to applying one’s knowledge to work.

AI inference, which I see as a significant part of the future labor market, is projected to be far more substantial than AI training.

We expect that a noteworthy part of the global expenditure on labor will gradually shift towards AI agents. These AI agents primarily inhabit cloud databases or data centers and function on specialized AI chips for training and inference.

If we hypothesize that AI companies could capture half of the labor cost savings at $25 trillion, and that half of that figure is spent on AI chips, we’re looking at a significant expense for AI companies. It’s these companies, including giants like Nvidia, that will likely receive a large portion of this investment due to their integral role in supplying the necessary hardware – AI chips – for training and inference activities.

The AI scaling laws could be the biggest finding in computer science since Moore’s Law was introduced. 📈 In my opinion, these laws haven’t gotten the attention they deserve (yet), even though they could show a clear way to make considerable improvements in artificial intelligence. This could change every industry in the world, and it’s a big deal.

👉 AI Scaling Laws – A Short Primer