Oracle and Anthropic Show the Next Phase of AI Contracting

AI buyers are no longer betting on one supplier; they’re locking in multi-vendor compute, financing, and infrastructure deals to secure long-term capacity.

Updated

What is this trend?

Large AI buyers are moving from single-supplier deals to multi-vendor, long-duration contracts that bundle chips, data centers, and financing to preserve supply optionality.

  • Oracle is broadening its stack with TPU, Cerebras, SambaNova, and Nvidia capacity.
  • OpenAI and Anthropic are locking in multiple suppliers across cloud, silicon, and leasing.
  • Procurement is becoming a financing decision, not just a capacity purchase.
  • Long-term obligations are concentrating power in vendors that can package compute, software, and capital.
  • Neoclouds face rising pressure as scale, funding, and independence become harder to sustain.

What’s the latest?

Oracle is now adding Google TPUs, Cerebras, and SambaNova while still scaling Nvidia capacity, extending the contracted-stack model into a more explicit multi-vendor procurement strategy.

Governance Gaps, Finance-Led Compute, and Multi-Vendor AI Contracting

How it developed

  1. Control moves up the stack, AI deals go contract-first, and sovereign compliance tightens
    • Compute Financing and Software Compatibility Are Hardening Nvidia’s Moat

Go deeper

Curated long-form picks on this trend — podcasts, videos, and analysis, by vantage.

Related trends

Stay ahead in Machine Learning

Get the weekly brief in your inbox — the developments, what they mean by vantage, and what to do next.