CRWV vs NBIS: AI Stock Comparison

• “In order to meet the demand for AI cloud infrastructure services, we and our partners may need to expand existing data centers through development projects or add new data centers through…” — 10-K risk factor (filed 2026-03-02) · source
• “If our data center providers fail to meet the requirements of our business, or if the data center facilities experience damage, interruption, or a security breach, our ability to provide…” — 10-K risk factor (filed 2026-03-02) · source

• “Technological developments in generative AI and inference, such as the development of AI models that require less computation power than earlier models, may result in decreased or differing…” — 20-F risk factor (filed 2026-04-30) · source
• “If there is insufficient customer demand to utilize the compute capacity we build, we would not realize the expected returns on our expansion efforts and our businesses, future operating…” — 20-F risk factor (filed 2026-04-30) · source

CoreWeave rents large NVIDIA GPU clusters to AI labs and hyperscalers. Results depend on contracted backlog, data-center delivery timelines and heavy debt-funded capex.

Nebius runs GPU cloud capacity in Europe, the U.S. and new markets. Recent capacity deals (NEOM framework Oct 21, Texas lease Oct 29) show its build-out pace.