This Mill Valley real estate deal is either a stroke of genius or a perfect symbol of how detached AI-era wealth has become from the rest of us — and honestly, it might be both.
Investment banker Storm Duncan is selling his 13-acre estate in Mill Valley, just north of San Francisco, and he doesn’t want your cash. He wants your Anthropic equity. Not a check. Not a wire transfer. Shares in one of the most talked-about AI companies on the planet. That’s the ask.
As someone who spends most of my time building bots and thinking about where AI infrastructure is actually headed, I find this deal fascinating — not just as a real estate curiosity, but as a signal about how AI company valuations are reshaping everything around them, including how people think about money itself.
When Equity Becomes a Currency
Anthropic’s valuation has reportedly grown from around $72 billion to a projected $135 billion by 2026. That kind of trajectory turns equity into something more liquid in people’s minds — not a stock you hold, but a currency you spend. Duncan is essentially betting that Anthropic shares are worth more than whatever cash a buyer might offer today.
That’s a bold position. Anthropic is not a public company. Its shares don’t trade on an open exchange. Accepting equity in a private AI firm instead of cash for a multi-million dollar property means Duncan is making a very specific bet: that Anthropic’s value keeps climbing, that a liquidity event happens, and that the timing works out in his favor.
From a bot builder’s perspective, I get the logic. The companies building foundational AI models — and Anthropic is squarely in that group with Claude — are sitting at a chokepoint in the industry. Every developer, every enterprise, every startup building on top of AI needs access to capable models. Anthropic has positioned itself as the safety-focused alternative in that space, and that positioning has attracted serious capital.
What This Says About the AI Bubble Question
There’s an uncomfortable question sitting underneath this deal. If the bubble pops — and plenty of smart people think AI valuations are running ahead of actual revenue — what happens to Duncan’s estate trade? He ends up holding equity in a company that may be worth a fraction of today’s projections, and the buyer walks away with 13 acres of prime Northern California land.
Real estate in Mill Valley doesn’t evaporate. Anthropic equity, in a worst-case scenario, could. That asymmetry is worth thinking about carefully.
But Duncan is an investment banker. He’s not naive about risk. The fact that he’s structured this deal suggests he genuinely believes Anthropic equity is undervalued relative to what it will be worth at exit. He’s not taking equity because he can’t find a cash buyer. He’s taking it because he wants it.
The Infrastructure Problem Nobody’s Solved Yet
One thing that keeps coming up in discussions about Anthropic’s future is its infrastructure challenge. Training and running large language models at scale is extraordinarily expensive. Anthropic hasn’t publicly laid out a clear path to solving that cost problem long-term. That’s not a knock — nobody in this space has fully cracked it — but it’s a real variable when you’re treating equity in the company like a savings account.
As someone building on top of these APIs daily, I see the cost pressure from the other side. Every token costs money. Every call to a model adds up. The companies that figure out how to deliver capable AI at lower infrastructure cost will have a serious advantage. Whether Anthropic gets there first, or whether a competitor does, shapes what that equity is actually worth.
A New Kind of Deal for a New Kind of Economy
What makes this story stick with me isn’t the property or even the dollar figures. It’s the idea that we’ve reached a point where AI company equity is being used as a medium of exchange for physical, tangible assets. A house. Land. Something you can stand on.
That’s a new wrinkle in how wealth moves around in tech-adjacent communities. And for those of us building in this space — writing code, designing bot architectures, shipping products on top of AI models — it’s a reminder that the financial layer of this industry is evolving just as fast as the technical one.
Whether Duncan’s bet pays off depends on factors neither he nor anyone else fully controls. But the fact that this deal exists at all tells you something real about where we are in the AI moment right now.
🕒 Published: