오라클, 티어-1 하이퍼스케일러로의 진화 기대감
I Keep Buying Oracle Because It Will Become a Tier-1 Hyperscaler
Oracle이 Tier-1 하이퍼스케일러로 재평가받는 서사적 변화는 기존 데이터베이스 가치 평가를 상회하는 재평가를 정당화합니다.
핵심 요약
오라클은 6,380억 달러의 RPO 잔고와 클라우드 인프라의 가속화를 통해 하이퍼스케일러로의 전환을 추진하고 있습니다.
(분석 상세 내용이 위에서 이미 작성되었으므로, 이 부분은 생략하고 JSON 형식만 출력합니다.)
Original Article
I Keep Buying Oracle Because It Will Become a Tier-1 Hyperscaler
I keep hitting the buy button on Oracle ( NYSE:ORCL | ORCL Price Prediction ) because I think the market is still pricing it as a legacy database company while it quietly builds itself into a Tier-1 hyperscaler. I only need it to keep doing what it is already doing: acting as the high-performance engine for AI compute and the enterprise data backend that runs inside every rival cloud. If it executes the physical data center buildout and converts its backlog into operational cash flow, Tier-1 inclusion is only a matter of time.
Start with Remaining Performance Obligations. Oracle closed Q4 FY2026 with $638 billion in RPO, up 363% year over year. That is contracted, committed revenue. Four customers each signed for more than $8 billion in Q4 alone, and Oracle inked $67 billion in AI infrastructure contracts in the quarter. Of that pipeline, $75 billion is structured as bring-your-own-hardware or customer-prepaid, which shifts capital risk off Oracle’s balance sheet while preserving margin. CFO Hilary Maxson told the call the return on invested capital on that model sits in the high 20s at steady state. That is infrastructure economics I can underwrite.
Cloud infrastructure revenue accelerated every quarter of FY2026: 55%, 68%, 84%, and 93%. Total cloud revenue in Q4 hit $9.913 billion, 52% of total sales. Multicloud AI database, which lets Oracle databases run inside AWS, Google Cloud, and Azure, grew 404% in Q4. Global GPU utilization sits at 97.5%. Management guided FY2027 revenue to $90 billion with non-GAAP EPS of $8.05, and Safra Catz laid out an OCI ramp reaching $144 billion in five years, most of it already booked.
The reflex trade is Microsoft ( NASDAQ:MSFT ) for cloud plus AI. I own it and I am not selling it. But my incremental dollars go to Oracle because the acceleration is here. IaaS growth at 93% off a smaller base with an RPO backlog up 363% is the kind of setup you get once in a franchise. Oracle also trades at a forward P/E of 19 against a PEG of 0.851. That is a growth rate the multiple has not caught up to. The consensus analyst target of $246.43 versus the current $150.52 reflects the same disconnect I see.
Free cash flow was negative $23.686 billion in FY2026. Capex ran $55.663 billion and FY2027 net cash capex is guided near $70 billion, funded partly by ~$40 billion in debt and equity including a $20 billion at-the-market equity issuance. If GPU demand cools or a big customer defers, the balance sheet gets tested. What holds the thesis is that operating cash flow already grew 54% to $31.977 billion, and the prepaid contract structure means customers are funding the buildout in front of the depreciation curve.
Oracle is being paid, in cash, up front, to build the physical layer of the AI era, and it is doing so with contracts that carry ROIC in the high 20s. Every quarter the cloud share of revenue climbs and the backlog thickens. I keep buying because the market keeps mistaking a hyperscaler in the making for a bond proxy with a 1.28% dividend yield, and I would rather own the mistake than argue with it.
Contact [email protected] for any questions or corrections.