About a year ago, a big corporate was pleading with its employees to adopt AI into their workflow. Today, the company has imposed a strict token limit.
This anecdotal scenario reveals the first cracks of a larger crisis. We are facing an ecosystem that, from the end consumer to the servers of tech giants, rests on fragile financial foundations.
The Value Paradox and the Rental Economy
To understand the problem at its core, we must abandon the comparison with the Industrial Revolution. In the past, steam engines made work faster and cheaper. The construction cost was high, but the factory owned the equipment. With each product manufactured, the unit cost decreased.
AI operates under different logic. Although it makes work more productive (where there is still controversy), with token-based pricing, you don't own the machine—you rent it, and each time it's used, the monthly bill grows in the same proportion. AI may reduce human labor time, but it can also raise the cost of use to a point where the supposed technological "replacement" becomes more expensive than the employee it was meant to replace.
Tokens are leverage. If this leverage is applied to routine work that doesn't generate increased revenue or cost reduction, using AI is not an investment—it's purely a new expense.
The Imminent Crisis
SaaS (Software as a Service) companies built on fixed monthly subscriptions are about to hit a wall. The current math is unsustainable: revenues are fixed (e.g., US$ 30 per month per user), but AI computing costs are exponential. The result is a cruel paradox: the software company begins to lose money exactly with its best and most engaged users. Growth will look great in usage charts, but the P&L (Profit and Loss Statement) will collapse.
The Logic Behind the AI Frenzy
If the financial equation of AI still doesn't add up, why does the market invest so aggressively? The answer combines suppliers' commercial strategy with corporate FOMO.
- The Dependency Strategy: AI suppliers subsidize technology costs to ensure mass adoption. The true price will only be charged after companies are already dependent on it.
- Structural Preparation: Companies know the game isn't about the present, but about the future. They are restructuring their infrastructure and culture now so that when AI reaches its peak, doing the work of hundreds of people, the house will already be in order.
-
The Survival Dilemma: For a CEO, embracing this dependency is a risk management matter dictated by Game Theory. They face only two scenarios:
- Scenario A (Invest and AI fails): The company loses money but absorbs the financial loss and survives.
- Scenario B (Don't invest and AI succeeds): The competition that bet on the technology will obliterate them in months, leading the company to bankruptcy.
``In summary: In today's corporate chess game, the cost of trying and failing is financial, but the cost of waiting and failing is fatal. It's the survival instinct that sustains the AI frenzy.
From SaaS to "Hype as a Service"
The scenario perfectly mirrors the collapse of the 2001 dot-com bubble, when telecommunications companies exchanged fiber-optic network capacities with each other just to register fake sales. The difference? Today, the AI loop is entirely legal under current accounting rules.
Market value giants today depend on one or two financially unstable startups. Speculative hype bubbles come and go, but they must eventually burst to force a more efficient reallocation of capital in the real world.
Major AI companies, however, try to prolong the bubble through dubious means. They generate wealth primarily through overvalued investor rounds rather than sustainable revenue. This has ceased to be "Software as a Service" and become "Hype as a Service".
If left unchecked, the AI Bubble risks being far more damaging than the dot-com one, not only in financial scale but by distorting global capital allocation and inflating pension funds with paper profits based on servers rented to themselves.
AI is genuinely powerful, but the market needs to stop buying narratives based on fear of the future and start demanding real profits in the present. The burst will be painful, but necessary.
