Oracle raises restructuring cost to $2.8bn amid AI data center spree

Oracle will spend an additional $700 million on its restructuring plan, the company disclosed in a regulatory filing on Friday. With the new amount, the total projected cost of the program for fiscal year 2026 rises to roughly $2.8 billion, up from about $2.1 billion set aside for restructuring in the prior fiscal year. According to Oracle, part of the new spending is tied to the adoption of artificial intelligence in some of the company's internal functions, as it reorganizes teams while trying to contain costs.

The price of the AI infrastructure race

The rising cost of layoffs coincides with an unprecedented expansion of Oracle's cloud infrastructure. Pressure for computing capacity grew after the company signed more than $30 billion in new AI-related cloud computing contracts in the first fiscal quarter alone, pushing its total contract backlog to $664 billion. The company estimates that roughly half of that backlog should convert into actual sales over the next 36 months. Despite the size of the contracts, Oracle says most of them will not require large additional capital outlays, because part of the deals involve upfront customer payments or equipment supplied by the clients themselves.

According to Bloomberg, in a filing sent to US regulators, Oracle calls the program its "2026 Restructuring Plan". Most of the cost, now estimated at about $2.8 billion, corresponds to severance paid to laid-off employees: $2.1 billion had already been provisioned, and the additional $700 million reflects, in the company's words, "additional measures expected to be taken". Oracle had already been cutting thousands of jobs this year to save cash, and Business Insider reported in August that the company was planning a new round of layoffs — part of the financial pressure created by its multibillion-dollar investment in AI data centers for clients such as OpenAI.

Cash under pressure despite record revenue

The numbers also show the other side of that expansion. In the first fiscal quarter, Oracle's revenue grew 30% year over year to $19.3 billion, beating market expectations. Even so, free cash flow was negative $5.4 billion for the period — a weaker result than usual for the company, though milder than the $9.56 billion cash burn analysts had projected. To sustain the expansion, Oracle said it plans to raise $40 billion in debt and equity issuance over the current fiscal year; of that total, $20 billion had already been raised through a stock sale in the first quarter.

• $2.8 billion — total projected cost of the 2026 restructuring plan
• $664 billion — Oracle's cloud computing contract backlog
• $40 billion — debt and equity the company plans to raise this fiscal year
• 21,000 — reduction in Oracle's headcount over one year

On a call with analysts, Oracle chief financial officer Hilary Maxson said "simplification and efficiency measures" should help lower costs and preserve margins. By the end of May, Oracle had about 49,000 employees in the United States and 92,000 abroad — 21,000 fewer than a year earlier. The cutback program is unfolding alongside moves at the top of the company: chairman Larry Ellison, who holds 40% of Oracle's shares, filed a new plan to sell up to 50 million shares by October 24, a batch worth roughly $8.75 billion at the late-June closing price.

What it changes for companies that depend on Oracle Cloud

For companies running AI workloads on Oracle Cloud Infrastructure (OCI), the supplier's financial strain is a signal to watch rather than an immediate alarm: record revenue shows demand for AI cloud remains strong, but negative cash flow and successive rounds of job cuts suggest Oracle is financing this expansion at the edge, with growing debt. That could translate into more frequent OCI price adjustments ahead, leaner support teams during the transition, and new data-center rollout timelines exposed to delays if access to capital becomes more expensive. Businesses betting their AI infrastructure on a single global vendor would do well to watch the company's coming quarterly results closely.