A New AI Milestone or Yet Another Stint? Data Center Investments Reach $61 bn in 2025
As Open AI floats through uncharted territory, could the $61 bn data center market actually reach profitability as promised?
ChatGPT now lets you adjust your email’s warmth levels. Alphabet acquired a new data center company. “The AI bubble is about to burst,” economists warn. Google announces new Gemini Flash 3 for speed. Everyone’s losing money on AI.
These are some of today’s headlines on AI. And they aren’t all enthusiastic. The response to AI has suddenly become quite diverse. And largely disappointing. It’s as if a veil has been removed, and the public perceives AI as more of the same high-level tech that’s supposed to cater to the chosen few.
Beyond this curtain? AI’s significance is dismissive.
However, that and countless warnings from economists haven’t stopped the AI enthusiasts. As the echo of the AI bubble burst makes the rounds every other day, another company ends up investing a few billion dollars in related infrastructure and hardware.
The disconnect is apparent.
The global data center market reached $61 billion this year. First, it was the chip frenzy that sent NVIDIA’s worth skyrocketing. And now, it’s the construction frenzy. The insatiable demand for AI isn’t nearly as evident as the demand for hardware, real estate, and energy. The nitty-gritty.
As an increasing number of data centers pop up, the market is questioning the returns. According to HBR, there are high variable spending, but low variable returns when it comes to AI.
The money movement is also apparent as all the tech and AI powerhouses hold hands to accelerate their AI roadmaps. It’s a well-thought-out strategy. But the returns are the real facet in question.
There’s not much to show.
Last week, the Wall Street Journal published a report on Notion. Its AI helps generate content, search, take down meeting notes, and research. It ate into 10% of Notion’s profit margin. And truly, it’s the actions that any user can carry out within meetings.
AI was equated with efficiency and cheaper labor costs. But it’s adding on- more than ever. Unproven returns. But enthusiastic overspending.
OpenAI will burn through approximately. $150 billion between 2024 and 2029, according to analysts. But it’s only in 2029 that the AI powerhouse could potentially turn a profit. Then it will have something to show for all its investments. To justify all the billions.
The global AI bubble may or may not pop, but investors and analysts have noticed a pattern-
The money movement is circular, and the entire US economy rests on that.
