The Unspoken Cost Shift of AI

The Productivity Paradox

Companies are celebrating massive productivity gains from AI tools while simultaneously laying off their expensive developers. But there's a problem they're not seeing: the math doesn't add up.

Whats not to love, you now don’t need to hire an expensive dev or engineer to create an idea. You can just vibe code your way to success. Look at this company. They were able to cut their development team more than in half by cutting “expensive” staff, purchasing AI tools and being more “productive” than ever before.

But then I do a quick search of Cursor (a popular AI vibe coding tool) and see this:

What I don’t think some business leaders are seeing is the actual cost of AI.

It was most clearly stated here:

and here:

What does this story tell us?

For the first time in history, we have shrunk the “I have an app idea” for development pipelines from idea to deployment without the “development”.

But we have automated it on a business structure that cannot thrive in 2025.

A Familiar Playbook

Remember Uber? Or AirBnB? Amazon? They blitzscaled their way to success. They elbowed their way into the market, grabbing at everything and anything that blinked their direction. They won.

They followed this business plan:

• Market Penetration - Price low to attract customers, subsidizing via VC firms and other financial tools.

• Market Consolidation - Drive out competitors who can’t hold a candle to the level of losses their VC firms can bear

• Market Extraction - Raise the price on customers to extract the most value while recouping their losses during the growth phase.

• Enshitification - I added this one, but make the product so bad, your customers hate it, but know there are no other competitors that can quite do what it can do so they are stuck with it.

The problem with doing blitzscaling in the AI market is companies are jumping on the cart before its been fully built.

Sure, you can now build things you never thought you could do as a single team. Sure these businesses have increased productivity in the short term, but they are laying off people, and building products on a business model that hasn’t been finalized yet.

The Subsidy Reality

AI businesses are still in the Market Penetration phase. Those savings you think you have with AI? They are subsidized by VC firms.

Soon those firms will come calling, and that AI tool you thought was so cheap and could replace humans will all of a sudden become less viable as an option because the cost will become astronomical. Look at how OpenAI is pricing their product. The pricing 'strategy' is literally guesswork. Sam Altman admitted as much earlier this year:

“I believe we tested two prices, $20 and $42,” he told the publication. “People thought $42 was a little too much. They were happy to pay $20. We picked $20. Probably it was late December of 2022 or early January. It was not a rigorous ‘hire someone and do a pricing study’ thing.”

You can also read more in my post 5 Predictions for 2025.

When the Music Stops

Don’t get me wrong, AI will have a positive impact on productivity and has definitely innovated on some issues and is really cool what it can do. But this is a cautionary tale that the “token pricing model” we will likely be moving toward once the technology moves to the market consolidation phase may end up costing more than what a human costs, and businesses will be left with a difficult decision: hire developers or pay the AI overlords and pass the cost to the customer.

That or the tech industry's compensation structure may come to resemble that of the retail sector.