The AI Revolution – Lessons from History and a Potential Shakeout

The artificial intelligence boom is reaching a critical juncture. After years of explosive growth, the economics of AI are shifting in ways that could upend who actually profits from the technology. While investors have poured billions into building the infrastructure behind AI, history suggests that the biggest winners may once again be the users—not the makers—of this new form of intelligence. Please see our initial blog on AI entitled, AI: The Intelligence Revolution

We may be reaching an inflection point where the baton passes from the companies that make AI possible to the companies that use AI to run their business more efficiently. As this transition happens, there are investment implications.

By 2X Wealth Group

The Bubble Building Beneath the Surface

The signs of excess are piling up. Oracle has taken on unprecedented levels of debt to fund its entry into AI infrastructure. Nvidia and OpenAI are effectively financing their own customers, creating circular flows of capital reminiscent of late-stage bubbles. Meanwhile, off-balance-sheet financing vehicles, called special purpose vehicles (SPVs) are obscuring the true scale of leverage across the sector.

It all feels familiar. During the telecom boom of the early 2000s, companies raced to lay fiber optic cables that would remain dark for years. Today, AI firms are building capacity that may never be fully utilized. Even government agencies are being asked to backstop private-sector bets, a sign that confidence may be giving way to concern.

The Open-Source Disruption

For years, the dominant assumption in AI has been that bigger models mean better performance. That scaling law is now being challenged. Researchers are finding that smarter algorithms and training techniques can outperform brute-force approaches that rely on ever-larger datasets and compute budgets.

The rise of open-source models—particularly from China—has accelerated this shift. The so-called “DeepSeek moment” has shown that models like Kimi2 Thinking can match the performance of leading proprietary systems at a fraction of the cost. With open models improving rapidly and available for free, innovation is spreading far beyond the handful of companies that once controlled the field.

Historical Parallels

Every major technological revolution follows a similar pattern: costs collapse as learning curves steepen. In AI, this process is happening at an unprecedented speed. As models train themselves more efficiently and hardware improves, the cost of processing tokens—the basic units of AI computation—is falling toward zero.

This dynamic echoes the lessons Warren Buffett drew from the early internet era. The technology changed the world, but the companies that built the infrastructure—Cisco, AOL, Yahoo—failed to capture lasting value. The same paradox may now be unfolding in AI: transformative impact, but disappointing returns for the builders.

The User Advantage

If history is any guide, the real winners will be those who use AI most effectively, not those who produce it. Asset-light companies like Apple can integrate AI into existing ecosystems without massive capital outlays. Data-rich platforms such as Google and Reddit can monetize user interactions more efficiently. In healthcare, robotics firms like Intuitive Surgical are enhancing precision and outcomes through AI-driven tools.

By contrast, the infrastructure providers face mounting risks. Their capital expenditures are soaring even as margins shrink. A single misstep—betting on the wrong chip architecture or overbuilding capacity—could leave billions stranded.

Managing the Cycle

Market psychology follows its own rhythm. Popping popcorn might provide a useful analogy. At first, just a few kernels pop — a few investors sell. As the heat rises, more follow. Eventually, nearly everyone sells in a cascade, though a few kernels never pop at all. That’s what bubbles look like in real time: early cracks trigger self-reinforcing fear, and once momentum shifts, it feeds on itself.

The fear of missing out (FOMO) drives the upswing. Ironically, the same fear — of being caught at the top — drives the fall.

For investors, the prudent approach may be to gradually reduce exposure rather than trying to time the exact peak. Dollar-cost averaging out of overheated positions can preserve upside potential while limiting downside risk.

Systemic Exposure

The AI narrative now dominates global markets. The ten largest U.S. stocks—many of them AI-linked—make up nearly 40% of the S&P 500’s total value. Consumer spending, corporate profits, and even GDP growth are increasingly tied to the fortunes of this single theme. If the AI trade falters, the ripple effects could be broad and deep.

The Paradox of Progress

Artificial intelligence will reshape industries, redefine productivity, and unlock new forms of creativity. But as with past technological revolutions, the financial rewards may not flow to those building the machines. They will flow to those who wield them best.In the end, intelligence—no matter how advanced—becomes a commodity. And as every investor eventually learns, the commodity always gets commoditized.

If you’d like help thinking through portfolio adjustments or risk management strategies, we’re here to help!

* * *

The material included herein is not to be reproduced or distributed to others without the Firm’s express written consent. This material is being provided for informational purposes, and is not intended to be a formal research report, a general guide to investing, or as a source of any specific investment recommendations and makes no implied or express recommendations concerning the manner in which any accounts should be handled. Any opinions expressed in this material are only current opinions and while the information contained is believed to be reliable there is no representation that it is accurate or complete and it should not be relied upon as such. Investing involves risk, including loss of principal, and no assurance can be given that a specific investment objective will be achieved.

The Firm accepts no liability for loss arising from the use of this material. However, Federal and state securities laws impose liabilities under certain circumstances on persons who act in good faith and nothing herein shall constitute a waiver or other limitation of any rights that an investor may have under Federal or state securities laws.

2X Wealth Group is a team at Ingalls & Snyder, LLC., 1325 Avenue of the Americas, New York, NY 10019-6066. If you would like to unsubscribe, please click here.