AI Is Extraordinary: Why Our Minds Struggle During Moments of Genuine Transformation
Introduction
Something unusual is happening in the markets right now, and you can feel it even before you can fully describe it. AI continues to accelerate the world in ways that feel both thrilling and disorienting—and Nvidia remains at the center of that acceleration.
Instead of one dramatic day of price action, we’ve now seen a series of swings that reflect the same deeper tension: bursts of AI-driven enthusiasm followed by abrupt pauses, as if the market keeps reminding us that momentum is not the same as permanence.
To make sense of this, it’s useful to introduce a framework from Daniel Kahneman’s Thinking, Fast and Slow, a landmark book that explains how we process information during uncertainty. Most people have heard the title of Kahneman’s book, but not everyone is familiar with the distinction itself, and it matters here. Thinking, Fast and Slow, one of the foundational works in behavioral economics. Kahneman describes two modes of thought:
System 1 — fast, emotional, intuitive, narrative-driven.
System 2 — slow, analytical, effortful, deliberate.
When markets move this quickly — and when a technology as powerful as AI captures our attention — these systems begin to pull in opposite directions. System 1 reacts to the pace of change and interprets it as inevitability. System 2 urges us to slow down and question whether the story and the timeline are truly the same.
Right now, System 1 is still shouting:
“Everything is changing—if you hesitate, you’ll miss the future.”
While System 2 remains the quieter voice reminding us:
“Technologies feel exponential long before their adoption becomes sustainable.”
This essay lives in the space between those two voices.
AI is extraordinary—that part hasn’t changed. But moments of genuine transformation remain the moments when our minds struggle most to interpret what we’re actually seeing.
Nvidia’s Earnings and the Early Infrastructure Cycle
Let’s begin with admiration—because Nvidia has earned it. Their execution is world-class. Their architecture is years ahead. Their ecosystem resembles a gravitational field more than a business moat.
But extraordinary earnings do not exempt a company from the realities of early infrastructure cycles. They intensify them.
Concentration among hyperscalers feels like strength until someone pauses spending. Pull-forward demand always appears permanent until digestion begins. And the AI incentive loop—where startups raise money to buy compute, then raise more because they bought compute—creates distortions that show up most clearly in cash flow timing.
These aren’t contradictions. They’re signatures of early-stage technological buildouts.
Why AI Will Change the World—Just Not on Today’s Timeline
It’s tempting to believe we’re already living in the final form of the AI economy. But technological revolutions don’t unfold in straight lines; they unfold in chapters.
The internet was inevitable.
Smartphones were inevitable.
Cloud computing was inevitable.
But inevitability didn’t prevent:
80% drawdowns,
multi-year sideways markets,
or valuation resets that punished even the winners.
Transformation is real.
But the timing of transformation is where humans—and markets—struggle most.
The Psychology of Technological Acceleration
Having spent my career evaluating how people process uncertainty, I’ve learned that intelligence does not immunize us against cognitive bias. In fact, in markets like this, intelligence often amplifies it.
Narrative Capture
When a story becomes big enough, we stop asking if it’s true and start asking why others don’t believe it yet.
Extrapolation Error
Exponential charts trigger a natural instinct to project the line forward forever.
FOMO Disguised as Insight
Momentum creates its own sense of urgency. What feels like analysis is often just fear of being left behind.
Confusing Technical Breakthrough with Market Maturity
AI’s technical curve is steep.
AI’s monetization curve is slow.
Eventually these curves meet. But never as quickly as the excitement implies.
Why Valuations Still Matter in Transformative Moments
When technology is extraordinary, valuation discipline becomes harder—not easier. Nvidia may dominate robotics, autonomy, healthcare, simulation, and industries not yet imagined. But even world-changing companies can become disappointing investments if expectations sprint too far ahead.
Cisco wasn’t wrong about the internet.
Intel wasn’t wrong about the PC.
Investors were wrong about the timeline.
Valuation is not cynicism—it’s patience.
The Cash Flow Debate Around Nvidia
For many investors, the most uncomfortable part of Nvidia’s story isn’t the revenue line—it’s the cash flow statement. On one side are critics pointing to surging accounts receivable, rising inventories, and the suggestion that “insane demand” isn’t showing up in cash as cleanly as it should. Some go further and argue that extended credit terms, strategic investments in AI companies, and ecosystem incentives amount to a kind of circular financing that will eventually collapse.
On the other side is Nvidia itself, along with a number of sober analysts, who frame the same data very differently. Management has been explicit that hyperscalers are buying on negotiated terms, not unusual vendor loans, and that the company is not secretly financing customers to prop up demand. They also point out that, even after this buildout phase, Nvidia is still generating very large amounts of free cash flow and sits on a substantial net cash position.
Both perspectives are responding to real signals in the numbers. Working capital has expanded sharply. Accounts receivable have climbed. Inventory tied to new architectures has grown ahead of shipments. At the same time, cash generation remains strong, and there is no single line item that, by itself, “proves” fraud or guarantees Nvidia’s demise.
The more useful way to read this, in my view, is not as a binary verdict but as a picture of where we are in the cycle. Early in a major infrastructure buildout, companies often act as a kind of shock absorber between aggressive customer demand, long production lead times, and shifting credit conditions. That produces noisy cash flow and uncomfortable charts long before it produces clarity.
Cash flow matters. But it is one lens among several: alongside customer concentration, competitive dynamics, the durability of AI workloads, and the eventual return on all of this capital spending. Reducing the entire Nvidia story to a single working-capital chart is just another way our minds try to force a complex, evolving situation into a simple, emotionally satisfying narrative—either “everything is fine” or “it’s all a house of cards.”
Cisco and Intel: Historical Echoes Worth Remembering
Are Cisco and Intel perfect analogues for Nvidia? No. But they are two of the clearest historical mirrors we have—not because the technologies are the same, but because the patterns are.
In the late 1990s, Cisco supplied the indispensable hardware that made the internet work. Demand felt infinite, revenues looked unstoppable, and the narrative became so powerful that valuations assumed exponential growth could continue without interruption. The internet didn’t fail—spending simply paused, infrastructure needed digestion, and Cisco’s valuation collapsed even while the company remained essential. The story was right; the timeline was wrong.
Intel offers a different echo. It sat at the gravitational center of personal computing, and for a time investors treated that dominance as a guarantee of perpetual growth. But PC demand proved cyclical, competition arrived, and the stock spent years going nowhere—not because Intel was unimportant, but because expectations had been pulled too far forward.
The point of these analogues isn’t that Nvidia will repeat their fate. Nvidia’s moat is wider and its platform more expansive. The point is that even world-changing technologies move in chapters, and markets often price the final chapter before the middle ones have played out.
Conclusion
AI is extraordinary. The world is shifting beneath us, and Nvidia may continue to redefine what is possible. But technological inevitability and market expectations operate on different clocks.
Our challenge isn’t to doubt the future—it’s to understand our own minds as we race toward it. Because in moments of genuine transformation, the most significant risk isn’t missing the future. It’s a misunderstanding of the present.
Author’s Note
I wrote this because I see the same tension in myself that I see in the market: the pull toward awe, excitement, and possibility, alongside the need for steadiness and perspective. AI is reshaping our world in real time, and none of us are immune to the psychological undertow it creates. My goal isn’t to dampen enthusiasm—it’s to widen the lens, so that we can hold both the wonder and the discipline at the same time.