Contents
What Oracle Is in 2026
Oracle Corporation is a 48-year-old enterprise software company that most of the technology industry spent a decade underestimating. Founded in 1977 by Larry Ellison, Bob Miner, and Ed Oates, Oracle built the relational database that became the backbone of global enterprise computing. It then spent 20 years watching cloud-native competitors like Google, Microsoft Azure, and AWS eat into its data center business.
In 2026 Oracle is no longer playing defense. Oracle Cloud Infrastructure (OCI) is the fastest-growing major hyperscaler by percentage revenue growth. Oracle is one of three founding partners in the $500 billion Stargate AI infrastructure consortium . The company signed more than 200 AI data center contracts in a 12-month window. And Larry Ellison is personally pitching heads of state on Oracle Cloud as the only infrastructure trustworthy enough to run national security AI.
Oracle reported $57.1 billion in FY2025 revenue, a 6% year-over-year increase, with cloud services and license support accounting for $40.5 billion of that total. The cloud infrastructure segment (OCI) grew at 52% year-over-year in its most recent quarter, significantly outpacing AWS (17%), Azure (29%), and Google Cloud (28%) on a relative growth basis.
OCI | The Hyperscaler Challenger
Oracle Cloud Infrastructure launched its second generation in 2018, built from scratch with a network architecture designed for the kind of low-latency, high-bandwidth GPU workloads that AI training demands. While AWS and Azure optimized for general-purpose cloud compute, Oracle engineered OCI around bare-metal GPU clusters with non-blocking RDMA fabric, giving it a measurable performance and price advantage for AI training jobs.
OCI is priced approximately 50% lower than equivalent AWS and Azure compute for comparable GPU instances. This cost differential has been a primary driver of Oracle winning workloads from AI labs, government agencies, and enterprises running large language model training and inference at scale.
| OCI Capability | Details |
|---|
Oracle's cloud region count, now exceeding 100, is the highest of any hyperscaler. The strategy is deliberate: governments and regulated industries that cannot send data across borders are a massive addressable market that AWS and Azure struggled to serve at scale. Oracle built dedicated sovereign cloud regions in the EU, Japan, Saudi Arabia, UAE, and multiple US federal classified environments.
Stargate | The $500B National AI Infrastructure Bet
In January 2026, Oracle joined OpenAI and SoftBank to announce Stargate, a joint venture to deploy $500 billion in AI infrastructure across the United States over four years. Oracle contributes cloud infrastructure and data center operations. OpenAI anchors the AI workloads. SoftBank provides capital and coordination.
The first tranche — $100 billion in immediate deployments — began construction in Texas. Oracle is building and operating the physical data centers, drawing on its existing OCI architecture. Larry Ellison framed the project in national security terms during the White House announcement, describing it as critical infrastructure equivalent to the interstate highway system.
Stargate gives Oracle something it never had before: a direct long-term contractual relationship with the most important AI lab in the world and a US government mandate to be part of the national AI infrastructure stack.
200+ AI Data Center Deals | The Pipeline No One Expected
Oracle disclosed in its FY2025 earnings calls that its remaining performance obligation (RPO) — essentially contracted future revenue — exceeded $130 billion, a figure that stunned analysts and confirmed that Oracle's AI data center pipeline was among the largest in the industry. The company signed more than 200 AI data center contracts in the 12 months ending mid-2025, with contract sizes ranging from hundreds of millions to multiple billions of dollars.
The customers span AI startups, pharmaceutical companies running drug discovery workloads, national governments, and Fortune 500 enterprises migrating legacy Oracle Database workloads to OCI. The stickiness of Oracle's existing database customer base gives OCI an advantage AWS and Azure cannot easily replicate: tens of thousands of companies already running Oracle databases have a natural on-ramp to OCI.
The Database Moat | 47 Years of Enterprise Lock-In
Oracle Database remains the dominant enterprise relational database for mission-critical workloads. Oracle holds an estimated 30%+ of the total database management system market by revenue, with its highest concentration in financial services, healthcare, government, and telecommunications — sectors where data integrity, ACID compliance, and support SLAs are non-negotiable.
Autonomous Database, Oracle's cloud-native database product, removes the need for a DBA by automatically patching, tuning, and scaling. It runs on OCI and represents Oracle's strategy to migrate its 400,000+ database customers from on-premises licenses to cloud subscriptions. The Autonomous Database had a 40%+ revenue growth rate in the most recent fiscal year.
Oracle's database moat is reinforced by switching costs that are among the highest in enterprise software. A large bank running 50TB of Oracle Database with custom PL/SQL procedures, Oracle Forms applications, and 20 years of stored procedures faces a multi-year, nine-figure migration project to move to PostgreSQL or another alternative. This lock-in has funded Oracle's cloud buildout.
Oracle Health | The Cerner Rebuild
Oracle acquired Cerner, the electronic health records (EHR) company, for $28.3 billion in June 2022, the largest acquisition in Oracle's history. The strategic rationale was clear: Cerner's EHR systems run in approximately 25% of US hospitals, giving Oracle a dataset of unparalleled scale for clinical AI and a direct line into the $4 trillion US healthcare market.
The integration has been turbulent. Oracle rebranded Cerner to Oracle Health, initiated a large-scale migration of hospitals from Cerner's on-premises infrastructure to OCI, and faced significant pushback from hospital IT departments over pace and contract terms. The US Department of Veterans Affairs — Oracle Health's largest single customer — issued multiple warnings about implementation delays in the first two years post-acquisition.
By 2026, Oracle Health has stabilized. The VA relationship is recovering. Oracle is positioning its clinical AI models — trained on de-identified data from Cerner's hospital network — as a differentiator against competitors like Epic. The health cloud division contributes an estimated $5-6 billion annually to Oracle's total revenue.
Financials | Revenue Breakdown 2025
| Revenue Segment | FY2025 Performance |
|---|
Larry Ellison | The Strategy Behind the Pivot
Larry Ellison, 80, has been Oracle's founder, chairman, and CTO since stepping down as CEO in 2014 — a role he gave to Safra Catz. He has become the public face of Oracle's AI and national security push with a conviction and urgency that surprised even longtime Oracle watchers.
Ellison's thesis is specific: AI will be the most important technology in human history, the data required to train and run AI systems cannot be trusted to foreign-controlled infrastructure, and Oracle's combination of sovereign cloud regions, database depth, and healthcare data makes it uniquely positioned to be the infrastructure provider for governments and regulated industries that will never put sensitive workloads on AWS or Azure.
He has personally met with heads of state in Saudi Arabia, Japan, the UAE, the UK, and India to pitch Oracle Cloud as national infrastructure. Multiple sovereign AI initiatives in those countries have contracted OCI. Ellison's personal involvement at the head-of-state level is a distribution channel no other hyperscaler replicates.
Ellison also holds a significant personal stake in Oracle — approximately 40% of shares outstanding — aligning his net worth directly with OCI's execution. His net worth crossed $200 billion in 2025, briefly making him the world's wealthiest person.
Competitors | Where Oracle Wins and Loses
| Competitor | Competitive Dynamic |
|---|
Oracle's competitive moat is unusual in the technology industry: it is simultaneously one of the most hated vendors in enterprise software (for aggressive license audits and lock-in practices) and one of the most indispensable. The companies that run Oracle systems tend to run them for decades.
In 2026 the question is whether Oracle can translate its database incumbency and OCI's cost advantage into hyperscaler-scale revenue growth before AWS and Azure close the GPU pricing gap. The $130 billion RPO suggests the AI data center pipeline is real. Whether OCI can execute at the build pace required to deliver it is the defining challenge for the next three years.
See Also
Discussion
Every comment appears live in our Discord server.
Join to see the full conversation and connect with the community.
Comments sync to our ObjectWire Discord · Oracle | Company Profile, Cloud, AI Strategy 2026.