Oracle stock hit a fresh 52-week low last Friday, and the shares are now down 62% from their all-time high, even as the company reported record-breaking demand for its cloud infrastructure services.
The drop reflects growing investor concern about whether Oracle's largest customers can actually pay for what they have ordered, and whether the company's heavy debt load is sustainable if some contracts fall through.
Oracle's data centers are built to serve the AI industry. They house graphics processing units from Nvidia and Advanced Micro Devices, and the company has constructed some of the largest GPU clusters available, allowing customers to run workloads using more than 131,000 chips simultaneously. Oracle also uses software-powered automation to bring new data centers online faster than competitors, and a proprietary networking technology that moves data between chips faster than traditional Ethernet systems.
That combination of speed and cost efficiency has attracted major AI customers. OpenAI, Elon Musk's xAI, Cohere, and Meta Platforms are among Oracle's top clients, according to Yahoo Finance.
The headline figure from Oracle's most recent earnings report was its remaining performance obligations, which jumped 363% to a record $638 billion. That number reflects signed contracts for services that have not yet been delivered. It functions like an order backlog and is widely used as a predictor of future revenue.
Oracle reported $19.2 billion in total revenue during its fiscal fourth quarter ended May 31, a 21% increase from the prior year period. Its cloud infrastructure segment brought in $5.8 billion of that total and grew at a pace of 93%.
But the size of the backlog has raised questions as well as excitement. The concern is that some customers may not be able to fulfill their commitments, which would be a serious problem because Oracle has taken on significant debt to build out new infrastructure to meet the demand. The Wall Street Journal previously reported that OpenAI alone might account for a substantial share of the backlog, and OpenAI's financial position has drawn scrutiny from analysts watching the AI sector.
The question for investors is whether the 52-week low represents a buying opportunity or a warning. The bull case rests on the scale and efficiency of Oracle's infrastructure, its growing roster of AI customers, and the long-term trajectory of demand for cloud computing power. The bear case centers on customer concentration risk, the debt taken on to fund expansion, and the possibility that AI spending by large customers could slow or shift.
The stock's 62% decline from its high has brought Oracle's valuation down considerably from levels that assumed flawless execution and continued backlog conversion. Whether the remaining performance obligations translate into actual revenue will likely determine the next move in the shares.
