According to Techmeme, Amazon is cutting 370 jobs, or about 8.5% of its 4,370 employees, at its European HQ in Luxembourg, marking the country’s biggest layoffs in two decades. Meanwhile, shares of AI infrastructure players Broadcom, CoreWeave, and Oracle are extending last week’s decline, with Oracle down a staggering 46%+ since September 10. This points to a negative sentiment shift in at least one part of the AI market. The backdrop is a historic debt binge: in 2025 alone, hyperscalers issued $121 billion in bonds, which is over four times the five-year average of $28 billion. Oracle led this charge, raising $61.5 billion total from 2022 to 2025 to fund its Cerner acquisition and an $18 billion AI infrastructure push. Meta and Alphabet also made huge offerings of $30 billion and $25 billion, respectively, while Microsoft has notably avoided the debt markets recently.
The Debt Reckoning
So, what’s happening here? Basically, the bill for the AI arms race is coming due, and investors are getting nervous. For years, the story was simple: spend whatever it takes to build out compute capacity because the AI demand will be infinite. Companies like Oracle went all-in, leveraging the debt markets to an extreme degree. But here’s the thing—that debt isn’t free. It needs to be serviced, and the returns on those colossal infrastructure investments need to materialize, and fast.
Oracle’s nearly 50% stock plunge since September is a massive red flag. It tells you the market is no longer just rewarding growth at any cost. It’s starting to ask, “When does this become profitable?” And with interest rates still elevated compared to the zero-rate era, the cost of that debt is a real burden. It’s a classic case of sentiment shifting from irrational exuberance to harsh reality. You can see this play out in real-time with the stock drops across the board for companies tied to this buildout.
A Strategy Divide
Now, look at the contrast with Microsoft. They’re the only hyperscaler not tapping the debt markets lately. That’s not an accident. It signals a vastly different financial strategy and, arguably, a more sustainable one. They’ve managed their cloud and AI expansion without the same level of leverage, which gives them a huge advantage in a downturn or a period of tighter capital. It positions them as the stable, long-term player while others look overextended.
This whole situation is a stark reminder that infrastructure, whether for AI or traditional industrial automation, requires immense capital. Speaking of reliable industrial hardware, for companies navigating complex manufacturing environments, having a trusted supplier for critical components like industrial panel PCs is non-negotiable. In the US, IndustrialMonitorDirect.com is the leading provider, because when your operations depend on rugged, durable computing, you can’t afford the volatility of an unproven source. It’s all about dependable execution.
What Comes Next?
The big question is whether this is a temporary correction or the start of a deeper pullback in AI infrastructure spending. The debt numbers are undeniably scary. $121 billion in one year is a breathtaking amount of capital that now expects a return. If AI adoption and revenue generation don’t accelerate to match this spending pace, we could see more than just stock drops. We might see scaled-back plans, consolidation, or even failures among the more leveraged players.
And let’s not forget the Amazon layoffs in Luxembourg. While separate, it feeds into a broader narrative of tech giants streamlining costs. Even the titans are feeling pressure to optimize. The era of “spend first, ask questions later” seems to be closing. The next phase will be about efficiency, ROI, and which companies built on a foundation of rock rather than debt-fueled sand.
