The AI Hype Tax: Are We Building Real Value?
A skeptical look at the rush to add AI onto every product — the cost of hype, the MLM-like money chain, content degradation loops, and what it takes to build real value.
I've been noticing something in our industry that's stuck with me: there's this rush to add AI everywhere, even when it doesn't make financial sense. A lot of companies are adding AI to products that work fine without it; not because customers are asking for it, but because everyone else seems to be doing it.
And I wonder if we're solving the wrong problem.
Few areas which keep me thinking....
The Real Force
Honestly, I think fear is behind most of these decisions. There's this underlying problem: "If we don't add AI, will customers leave us for someone who does?" It feels more reactive than intentional.
I keep thinking about AI's original promise—that it would help us tackle meaningful problems faster, that it'd let engineers focus on what actually mattered. Somewhere along the way, that shifted. Now I see AI getting added to products that don't really need it, creating a lot of noise without much signal.
What worries me most: we're spending three dollars to make one dollar back. And someone's paying for that gap. The clear winners are OpenAI, Anthropic, Nvidia. Everyone else is kind of gambling.
The Job Anxiety
There's a lot of fear about AI replacing jobs, which I think is making this worse. But wasn't the idea that AI would free us up to do more meaningful work? To move past repetitive tasks?
Instead, I'm seeing the opposite. The narrative has become about replacement and cutting costs. We're using it to reduce headcount instead of to help people solve harder problems together. That part troubles me.
The Money Question
When I trace the money, I notice something:
- Your customer pays you
- You pay the cloud provider
- They pay the LLM vendor
- Everyone pays Nvidia
There's value being created somewhere in that chain, but I'm not always sure where.
The Data Problem
Here's something that doesn't get enough attention: everyone's using free LLMs to generate content at massive scale. It sounds efficient, but there's a hidden cost.
All that AI-generated content gets scraped back into training data for the next model. We're creating this feedback loop where lower-quality output trains the next generation, which produces even lower-quality output. The internet is slowly becoming detached from original human thinking.
The End Game
Here's what concerns me: if we outsource too much thinking to these systems, what happens to our ability to think deeply?
Over the time, we'll have lost something important. The next generation won't develop the same problem-solving skills that come from struggling with hard problems. And when these systems fail or become too expensive, we won't have that foundation to rely on.
Closure
I think the companies that survive this era are the ones solving genuine problems.
A healthcare company using AI to predict patient deterioration before it happens? That saves lives and reduces costs. That's real value.
A project management tool adding an AI assistant when the actual problem is a confusing UI? That's just noise.
One solves something people actually need. The other is checking a box.
When the hype eventually cools, the companies still standing will be the ones that built on something real. The others will be forgotten.
The most powerful technology is still human thought.