According to the latest reports, Microsoft has significantly increased capital expenditures in recent quarters on data centers and computing infrastructure needed to support AI models, including solutions tied to OpenAI and Copilot. In fiscal year 2024 alone, the company’s capex exceeded USD 55 billion, with forecasts for the coming year pointing to further growth—driven primarily by demand for cloud services and generative AI. This represents a sharp increase compared to roughly USD 44 billion spent two years earlier.
While Microsoft’s revenues continue to rise, AI-related operating costs are growing faster than sales. The maintenance and expansion of data centers, the purchase of specialized chips (including GPUs for model training and inference), and electricity consumption have become some of the company’s largest cost items. In its latest financial report, Microsoft signaled that operating margins in the Intelligent Cloud segment are under pressure, despite solid revenue growth at Azure.
In this context, analysts believe Microsoft may turn to significant workforce reductions as a cost-control measure. The company currently employs around 221,000 people worldwide, and previous layoff rounds—most notably in 2023—affected roughly 10,000 employees, or close to 5% of the workforce. Analysts suggest that the January cuts could be of a similar scale, although the company has not yet confirmed any specific numbers.
Cost pressure is intensifying at the same time Microsoft is aggressively monetizing AI across its product portfolio. Corporate Copilot subscriptions are priced at USD 30 per user per month, but analysts note that at current pricing levels, AI profitability remains limited—especially given the high consumption of computing power. As a result, rising AI revenues do not necessarily translate directly into proportional profit growth.
If Microsoft ultimately moves forward with major layoffs in January, it will be another signal that the AI boom does not automatically guarantee job stability. On the contrary, soaring infrastructure costs and heavy investment requirements may force even the largest technology players to cut jobs, despite record spending on AI—simply because those investments may take longer than expected to pay off.
