The 1990s called – they recognize the vibe.
Walking through the current AI landscape feels like a massive case of déjà vu. It reminds me of the early days of the internet. Back then, we knew something tectonic was shifting, but no one—not the experts, not the CEOs, not the pundits—truly predicted how it would fundamentally rewrite the DNA of every industry.
We also didn’t predict the “Great Consolidation”—the way thousands of experimental firms eventually collapsed into a handful of giants. Today, we are seeing the same pattern: a whirlwind of hype, “vibe laundering” (turning vague excitement into fake financial certainty), and a lot of people pretending they have a crystal ball.
The “Nobody Knows Anything” Era In the film industry, William Goldman famously said, “Nobody knows anything.” That is the exact state of AI today.
- The Intelligence Crisis: We are flooded with AI-generated noise, making it harder to find “alpha” or objective truth.
- The Prediction Trap: Investors are trying to guess which model will “win,” yet we are still in the “spending phase.” The “earning phase” remains a theoretical vibe.
Why I’m Not Picking a Winner I’m frequently asked which AI model is “better” or which company will dominate the next decade. My answer? I don’t know, and I don’t think it matters yet.
Just as the early internet was about the thrill of a 56k modem and the first chat rooms, the current state of AI is about the infrastructure and the interface. I’m moving the conversation away from “Magical AI” (the hype) and toward “Functional AI” (how we actually use it).
My Strategy: Radical Honesty and Play Instead of waiting for the dust to settle or pretending to be a prophet, I’m leaning into “The Great Unknown.” Here is how I’m navigating the shift:
- Embracing the Sandbox: I’m not here to crown a king; I’m here to play. I love testing these tools, breaking them, and seeing where they actually add value versus where they just generate fluff.
- Focusing on the “Hardware” of Work: While software outcomes are uncertain, the way we build—our data architecture, our hardware needs, and our human-in-the-loop oversight—is measurable and real.
- Building for Flexibility: Because I know the market will likely consolidate (just like the internet did), I’m building workflows that are model-agnostic. I don’t marry the tool; I marry the process.
The Bottom Line We are living through a “Global Intelligence Crisis,” where signal is rare and noise is everywhere. The brands and individuals who will survive aren’t the ones making the loudest predictions—they are the ones being radically honest about the limitations of the tech while staying curious enough to keep playing.
The future of AI is unwritten. Let’s stop trying to predict the ending and start enjoying the experimentation.