Hype is not free.
It taxes executive attention. It taxes the credibility of the people who later have to implement something real. It taxes the next initiative, because the last one promised the moon and delivered a slide deck.
I am not interested in being the loudest voice in the room. I am interested in being the most useful.
What the tax looks like
In the last few years, a lot of companies paid the hype tax in full:
- Strategy offsites that produced "AI transformation" language without a single owned workflow
- Tool sprawl: six copilots, zero standards
- Creative teams demoralized by executives who believed "the AI can just do it"
- Risk teams shut out until a customer-facing incident forced the conversation
The Stanford HAI AI Index is useful for separating capability progress from marketing narrative. Capability is real. Theater is optional.
Useful sounds like: "This use case is strong. Start here." Or: "This use case is theater. Skip it." Or: "This vendor deck is three years ahead of your data reality."
If that feels less exciting than a keynote, good. Excitement is a poor operating system for capital allocation.
How I brief leadership rooms
When I speak to leadership teams and industry groups, I use a simple three-bucket frame:
- Now: High signal, human-reviewed, measurable within a quarter
- Next: Requires data or process work before tools
- Never (for us): Hype-aligned, brand-risk heavy, or pure status theater
Most rooms over-index on "Next" and "Never" while calling them "Now." That is the tax.
Pay the tax down
Before you fund the next AI initiative, do the unglamorous cleanup:
- Kill one zombie pilot this month
- Publish an internal "we will not automate this" list
- Require a metric and an owner on every AI proposal over a fixed spend threshold
- Bring legal, brand, and ops into the same room before procurement
The organizations that win the next five years will not be the ones who adopted the most models. They will be the ones who paid the least hype tax per dollar of real throughput.