Summary
A top executive at Nvidia recently shared a surprising fact about the current state of technology: running artificial intelligence is often more expensive than paying human workers. While many people fear that AI will quickly replace jobs, the high cost of computing power is slowing down this shift. Even as big tech companies announce large layoffs, experts say the move to AI is driven more by high investment costs than by the technology being a cheaper alternative to people right now.
Main Impact
The biggest impact of this trend is a shift in how companies manage their money. Instead of saving money by using AI, many businesses are finding that AI tools are "blowing away" their budgets. This means that while some jobs are being cut, the money saved is not staying in the bank. Instead, it is being spent on massive data centers, specialized computer chips, and huge amounts of electricity. This creates a strange situation where companies are smaller in terms of staff but are spending more money than ever before on technology.
Key Details
What Happened
Bryan Catanzaro, a vice president at Nvidia, explained that for his team, the cost of "compute"—which refers to the processing power needed to run AI—is much higher than the cost of his employees. This statement highlights a major hurdle for the AI industry. While AI can do many tasks, the hardware and energy required to perform those tasks at a high level are still very pricey. This makes it hard for businesses to justify replacing humans with machines for every task.
Important Numbers and Facts
Several reports and studies support the idea that humans are still the more affordable choice for most work. An MIT study found that AI is only cost-effective in about 23% of jobs that involve visual tasks. In the other 77% of cases, it is still cheaper to pay a person. Despite these costs, spending on AI is reaching record levels. Morgan Stanley reports that big tech firms plan to spend $740 billion on AI infrastructure this year alone. This is a 69% increase from the previous year. Meanwhile, the tech industry has seen over 92,000 layoffs in the first few months of 2026, showing a clear move toward spending on machines over people.
Background and Context
For a long time, the common belief was that AI would help companies save money by doing the work of many people for a low cost. However, the reality is more complicated. AI models require thousands of powerful chips and a constant supply of energy to stay running. Additionally, many AI software companies are currently losing money. They often charge a flat monthly fee, but the actual cost of running the AI for a heavy user can be much higher than what the user pays in their subscription. This has led to a "mismatch" where the technology is growing fast, but the business side is still trying to catch up.
Public or Industry Reaction
Leaders in the tech world are starting to speak out about these rising costs. The Chief Technology Officer of Uber recently mentioned that he had to rethink his entire budget because AI coding tools were costing much more than expected. Financial experts also note that AI software fees have jumped by 20% to 37% in just one year. This has caused some companies to view AI as a "helper tool" rather than a full replacement for workers. They are waiting for the technology to become more efficient before they fully commit to using it for everything.
What This Means Going Forward
The cost of AI is expected to drop significantly in the coming years. Experts predict that by 2030, the cost of running large AI models could fall by as much as 90%. As the hardware becomes better and more energy-efficient, the "tipping point" will arrive. This is the moment when AI becomes both cheaper and more reliable than human labor. Companies will also likely change how they charge for these tools, moving away from flat fees to charging based on how much the AI is actually used. For workers, this means there is a short window of time where human labor remains the more economical choice, but that window is slowly closing.
Final Take
Right now, the high price of technology is acting as a shield for many jobs. While AI is powerful, it is not yet a bargain for most businesses. The current wave of layoffs in the tech sector is more about companies shifting their wealth into expensive hardware rather than AI being able to do every job perfectly. The future of work will depend on how quickly the costs of computing fall and how reliable these systems become at a large scale.
Frequently Asked Questions
Why is AI more expensive than human workers?
AI requires a lot of expensive computer chips, massive data centers, and a huge amount of electricity to run. These "compute" costs currently add up to more than the salaries of many employees.
Are companies still hiring people despite the AI boom?
While many big tech companies are laying off workers to fund AI projects, humans are still cheaper for about 77% of tasks. Many businesses still rely on people because they are more cost-effective and predictable.
Will AI ever become cheaper than humans?
Yes, experts believe the cost of running AI will drop by 90% by the year 2030. As the technology becomes more efficient and the hardware becomes cheaper to make, it will likely become more affordable than human labor for many more roles.
