Oracle · one lens
Benjamin Graham on Oracle
Oracle’s cloud contracts may become valuable, but its cash outflow and rising debt leave Graham’s margin of safety unmeasured.
Oracle closed at $141.40 on October 9, 2026, up 4.21%, to $141.40 from the previous close of $135.69, per the supplied market data. Under Graham’s Mr. Market test, Oracle must treat that quote as an offer, not a valuation. His investment test requires thorough analysis to support safety of principal and an adequate return. Oracle needs a defensible estimate of intrinsic value before a margin of safety can be measured against its price.
Oracle spent $55.7 billion on capital expenditures in fiscal 2026, compared with $21.2 billion in fiscal 2025, per the company dossier. Oracle generated negative free cash flow of $23.7 billion in fiscal 2026, compared with negative $394 million in fiscal 2025 and positive $11.8 billion in fiscal 2024. Oracle ended fiscal 2026 with $156.2 billion of total debt, compared with $104.1 billion a year earlier. Oracle cannot demonstrate Graham’s defensive test of conservative financing from those figures. The supplied record also lacks the earnings history needed to check his price ceilings: 25 times average earnings over seven years, versus 20 times earnings over the latest twelve months.
The Wall Street Journal reported an OpenAI computing contract of roughly $300 billion over roughly five years, against a previously disclosed contract of roughly $30 billion a year; those contracts have different terms and are not directly comparable. The dossier says Oracle expected the vast majority of newly disclosed revenue to come from OpenAI. The Financial Times reported a Tencent lease too, so OpenAI is not the only customer in this record. The Jupiter report describes power and permitting hurdles at a data center to which Oracle made financial commitments. Graham’s test makes the gap between contracted revenue and cash collected after construction costs decisive.
The October financing report says Oracle was in talks to finance a large chip purchase; the report does not establish completed financing. The CNBC report describes a lower estimate of OpenAI’s annualized revenue than previously reported, not a failure to pay Oracle. Per the TEXXR coverage record, Oracle appeared in 27 articles in calendar 2026’s third quarter, the latest complete quarter, against an eight-quarter average of 20 articles per quarter. TEXXR measures attention with that count, not contract performance. The supplied record gives Oracle no demonstrated path from its commitments to cash after spending and financing costs. Oracle’s lower price alone therefore cannot establish a margin of safety.
An editorial application of Benjamin Graham's published framework — not their actual view, affiliation, or advice.
Where this lens comes from
Benjamin Graham's framework is set out at the lens page, drawn from The Intelligent Investor, Security Analysis. This page applies it to Oracle and nothing else — the company's full record is in the dossier.