If you only read the headlines, you see a simple, scary story: “Company X lays off workers because of AI.” The first thing that comes to mind is that algorithms are taking over jobs that people used to do. It’s a story about things becoming useless.
I think this interpretation is not only too early, but it also hides a bigger and more strategic business truth. The truth is that AI is not taking over many jobs right now. We are in a time of strategic capital reallocation.
The truth is that AI isn’t doing the work; companies are making a tough choice: they are cutting jobs to pay for the AI they think they need to stay in business.
The Staggering Cost of the AI Revolution
You need to know that building AI for businesses is not the same as using a new SaaS tool to understand these choices. This is a heavily capital-intensive industry. The launch costs are astronomical. Think about what you need:
- Specialized Hardware: Getting thousands of GPUs and custom accelerators, which are hard to find and expensive.
- Physical Infrastructure: Building or updating data centers with many servers to handle the workload.
- A lot of energy is used: these systems need a constant, huge amount of electricity to run.
- Elite Talent: In a very competitive global market, finding and keeping the best AI researchers and engineers.
The numbers are mind-boggling. Training just one state-of-the-art model can cost tens of millions to hundreds of millions of dollars. The time of the billion-dollar training run is coming up fast. This is before the costs of safety, compliance, deployment, and product integration. The idea isn’t just a theory. Check out the moves from the front lines:
- Microsoft is spending tens of billions of dollars on AI infrastructure and has told investors that the expenditure is hurting their profits. They also cut thousands of jobs, saying it was a necessary restructuring to make the business more AI-friendly. (Microsoft Investor Relations)
- Amazon cut thousands of jobs in areas like Alexa and Prime Video, but at the same time, it announced a giant expansion of AI data centers and services like AWS Bedrock.
- Autodesk cut its workforce by 9% and said the move was to “focus resources on AI and platform priorities.” The language from Intel and others has been nearly identical.
This planned change is not a coincidence. It’s a planned transfer of money from some parts of the business to the new AI frontier.
The Critical Misunderstanding: Replacement vs. Reallocation
Most people who write about this don’t get this important point. Usually, when companies lay off workers, they don’t say, “We have an AI that can now do the work of these 1,000 people.” Memoranda and investor calls convey a distinct narrative. The words used are about getting rid of “legacy” or “non-core” product lines or making operations more efficient and cutting down on management layers or even reallocating resources to focus on higher-growth priorities. To put it simply, we are taking apart one cost center to build and pay for a new, huge one.
AI is starting to take over some jobs in customer service, content creation, and code testing, but we aren’t seeing a big wave of job replacement on a macroeconomic level. The layoffs right now have less to do with what AI can do today and more to do with what AI will be able to do in the future. It’s a colder, more businesslike decision. The business has decided that it can’t afford both the workers it has now and the AI future it wants. So, it picks the future.
The Leadership Pressures That Are Causing This Change
What motivates leaders to make these challenging decisions that may be subject to public criticism? Three factors are creating a strong demand:
- A lot of pressure from investors: Public markets are giving companies with a strong, believable AI story more money than they deserve. Telling the market that you are making a big investment in AI, with money saved from other areas, is a strong sign. It says, “We are serious, focused, and ready to make hard choices.” Conversely, businesses that exhibit uncertainty about AI face penalties. In this case, layoffs are a strange way to show commitment.
- The Unavoidable Margin Squeeze: AI costs so much that even the most profitable companies can’t just add it to their other costs without hurting their margins. The most common financial logic is to trade costs instead of stacking them. It is important to identify funding sources, and for most tech companies, payroll represents the largest expense.
- The Existential Strategic Repositioning: It’s not just about getting things done. Big Tech is going through a major change. They don’t want to just use AI anymore; they want to be the platform that all other companies use to access AI. This means that resources need to be used in a different way: fewer bets on products from the past and all-in bets on AI platforms of the future.
We need to change the subject. The current wave of “AI layoffs” is mostly a way to get money. It’s moving money from people to technology.
The story of direct replacement is a distraction from a more complicated truth: we are paying for a new industrial revolution, and the money is coming from our paychecks. This is a strategic decision, not a technological necessity. I explore what this new industrial revolution means for the future of work in my Life in the Digital Bubble series.
What do you think? Is this capital reallocation a necessary evil to stay competitive or a risky short-term plan? If your company is facing similar decisions and you want a structured way to navigate AI investments, roles, and risks, you can contact me through the Consulting page.