Tom Lee Calls Bottom on AI Stock Selloff, Citing Tight Supply and Crypto Resilience

The Market Correction Trap

Lately, it feels like the honeymoon phase of the AI gold rush has hit a wall. We have seen significant selloffs in high-flying tech stocks, and the usual suspects in the financial media are quick to declare the end of an era. But if you look at the mechanics of what is actually happening, the narrative of a bubble bursting does not quite hold up. Fundstrat’s Tom Lee is currently arguing that we have hit a bottom, and while I am usually the first person to roll my eyes at a bottom-caller, his reasoning deserves a closer look from a founder’s perspective.

We are not seeing a lack of interest in artificial intelligence. Instead, we are seeing the physical and financial limits of how fast this technology can actually be deployed. For those of us building in this space, this is a crucial distinction. The selloff is not a rejection of the technology; it is a realization that the infrastructure cannot keep up with the hype. When supply is tight and growth slows, the market panics, but the underlying utility remains unchanged.

The Supply Chain Bottleneck

The primary reason for the recent cooling in AI stocks is not a lack of customers. It is a lack of chips, energy, and data center capacity. We have reached a point where even the wealthiest tech giants are waiting in line for hardware. This creates a natural ceiling on growth rates. If you cannot get the H100s, you cannot train the models. If you cannot train the models, you cannot ship the features that justify a trillion-dollar valuation.

Lee’s point is that these constraints have artificially suppressed stock prices. We have seen a massive drawdown, but the demand for compute is still essentially infinite. For a builder, this is actually good news. A cooling market flushes out the tourists and the "wrapper" startups that were just riding the momentum. It forces the industry to focus on efficiency. If compute is expensive and hard to come by, you stop building useless chatbots and start focusing on high-margin, high-impact applications.

Crypto as a Fear Barometer

One of the most interesting parts of Lee’s thesis involves the resilience of the crypto markets. Traditionally, when tech stocks take a bath, crypto follows suit, often falling twice as hard. But we have seen something different recently. Even in the face of significant negative news—like the recent security issues at Binance—the market did not spiral. It absorbed the shock and held its ground.

To an analyst, this suggests seller exhaustion. The people who were going to panic-sell have already left the building. When a market stops reacting to bad news, it usually means the bottom is in. For crypto founders, this is a sign that the "weak hands" have been replaced by long-term participants who are less concerned with daily price action and more concerned with the actual utility of the protocols.

Why This Matters for Founders

If Lee is right and we are at the bottom of this AI selloff, the next phase will not be about hype—it will be about execution. The initial rush is over. The "free money" phase where you could raise a seed round with a slide deck and the word "transformer" is closing. What comes next is the hard work of building sustainable systems that work around the supply constraints we currently face.

We are seeing a shift from general-purpose AI toward specialized, vertically integrated solutions. If you cannot get enough compute to train a model that knows everything, you train a model that knows one thing perfectly. This is where the real value will be created in the coming year. The market correction is essentially a filter, separating the teams that can navigate resource scarcity from the teams that only know how to spend venture capital.

Navigating the Noise

It is easy to get distracted by the volatility. When you see your net worth or your company’s theoretical valuation swing by 20% in a week, the instinct is to pivot or hibernate. But looking at the resilience in crypto and the persistent demand in AI, the signal is clear: the fundamentals are still there. The world still wants decentralized finance and automated intelligence; it just needs them to be reliable and scalable.

The selloff in AI stocks was a necessary reality check. The market was pricing in a future where growth was linear and infinite, ignoring the fact that we live in a world of physical hardware and finite electricity. Now that the expectations have been reset, we can get back to building things that actually matter.

The market correction is not a sign of failure; it is a sign of maturity. We are moving from the era of experimentation to the era of implementation.

The Takeaway for Builders

Do not let the headlines dictate your roadmap. If you are building in AI, focus on efficiency and solving the supply problem. Whether that means optimizing local inference or finding ways to use less data, the winners will be those who can do more with less. In the crypto space, take the market’s resilience as a vote of confidence. The infrastructure is holding up, and the audience is still here.

Tom Lee’s call for a bottom might be right, or it might be slightly early, but the underlying observation is solid. The panic is subsiding, and the builders who stayed focused during the dip are the ones who will lead the next cycle. Stop watching the charts and start looking at your code. The floor is established; now it is time to build the next level.

Read the original at Decrypt →