Oracle stock could hit $400, but it's still not a top AI infrastructure pick

Oracle delivered better-than-expected fiscal Q1 results and guided for at least $90 billion of revenue and $8.10 in adjusted EPS for fiscal 2027, both above Street expectations. Despite the positive outlook, Jefferies analyst Brent Thill said Oracle is not among his top three AI-infrastructure picks, tempering the investment-case enthusiasm.

Analysis

ORCL’s valuation debate is no longer about whether demand exists; it is whether the company can convert backlog into revenue without materially diluting infrastructure returns. The 2027 framework implies a steep ramp in OCI capacity, so the next 1-3 months should hinge on disclosures around remaining performance obligations, capex, GPU availability, and the mix of customer-funded versus Oracle-funded buildouts. A revenue beat alone is insufficient if capex rises faster than operating cash flow, because AI infrastructure multiples compress quickly when FCF conversion disappoints.

Competitive positioning is more nuanced than the headline enthusiasm suggests. ORCL can win workloads requiring dedicated capacity, database adjacency, and multi-cloud deployment, but hyperscalers MSFT, AMZN, and GOOGL retain distribution advantages and can subsidize AI infrastructure through broader cloud ecosystems. The second-order beneficiary is NVDA if Oracle’s capacity commitments are incremental rather than reallocated; conversely, suppliers such as AVGO and networking/power names benefit only if Oracle confirms sustained deployment cadence rather than a one-off cluster build.

Consensus may be underestimating the financing and execution risk embedded in a rapid OCI expansion. Large customer contracts can create impressive RPO optics while concentrating credit, renewal, and utilization risk; a delay in data-center power availability or customer go-live dates would push revenue recognition out while depreciation begins immediately. Over 6-18 months, the key falsifier is not the 2027 target but whether incremental cloud gross margin and FCF margin stabilize as capacity comes online; sustained negative FCF or another material capex step-up would challenge the premium AI narrative.

Key Decisions for Investors

• Maintain ORCL as a tactical long only into evidence of funded capacity conversion over the next 1-2 earnings reports; add after confirmation that OCI growth is accompanied by stable-to-improving cloud gross margin and operating cash flow. Exit or hedge if capex guidance rises materially without a corresponding increase in contracted backlog or revenue timing.
• Express relative AI-infrastructure exposure via long NVDA / short ORCL in equal dollar risk for 3-6 months if ORCL’s post-results move outpaces NVDA: NVDA captures broad-based accelerator demand, while ORCL bears customer concentration, data-center execution, and depreciation risk. Cover if Oracle discloses multi-year take-or-pay commitments sufficient to de-risk utilization.
• Watch ORCL’s next filing for RPO growth, top-customer concentration, capex-to-revenue, and free-cash-flow conversion; do not initiate a standalone bearish position absent deterioration in these metrics, since confirmed large-cloud contracts can sustain momentum despite near-term cash-flow pressure.
• For a bullish ORCL position, use defined-risk call spreads dated 6-9 months rather than outright calls: upside requires multiple earnings prints validating the OCI ramp, while a capex-driven FCF miss can produce a sharp multiple reset before the 2027 revenue target is relevant.