Thirty years ago, very few businesses and services built atop Industrial Revolution assumptions inherent in mass employment, mass housing, mass media, and mass markets could pivot to the Internet. A place built on oppositional assumptions clustered around long-tails, niche tastes, small audiences, and rabid tribes.

Now, we have transitioned as many businesses and services as possible to the current Internet-driven paradigm. There are still remnants of businesses and services built on the old Industrial Revolution model, and there is still rampant emotional and cultural nostalgia for the safety and seeming solidity of “the good old days.” But by and large, the assumptions of the Internet continue their relentless march across our markets, our behaviors, our psyches, and our assumptions.

The current hype cycle around large-language models and machine learning is largely similar in tone, emotional content, and perspective to the previous Internet hype cycle that existed back when Pets.com was advertising opposite the Super Bowl around 1999 or so. Just as then, we currently are so early with our large-language models that we don’t have a “use case” for productivity, service, entertainment, or markets that seems to justify the current LLM hype cycle. Just as we didn’t have use cases at the arrival of the Internet.

But the use cases will arrive.

The hype cycle will collapse.

And out of the trillion dollars of hype cycle rubble left behind (it was billions after the dot-com crash in 2000-2002), businesses and services that were built on the assumptions of niche, long-tails, and tribalism will either adapt or expire. Even as new assumptions–LLM-driven assumptions–will begin their thirty-year-long takeover of our markets, our behaviors, our psyches, and our assumptions. Rest assured, it will take about thirty years for the takeover to even be 85% complete. No matter how hot the current friction is right now around LLMs.

When humans figure out how to make a technology work, when they stumble on an appropriate use case, they move like lightning, and all of us wonder how we lived without it. The fascinating element in all of this is who, exactly, will see the future use case and have the courage to go first.