The Enterprise AI Platform War Is Over. Now What?

A year ago, every consulting firm I knew was hedging. Run workloads on every model, see what sticks, tell clients you're platform-agnostic. It felt strategic. It might have been fear...

The market just made a decision. Anthropic passed OpenAI in business AI subscription spending, 41% to 39.5%, and the gap is growing. Meanwhile, OpenAI is offering the U.S. government a 5% equity stake as it pivots toward an IPO and government contracts. These two companies didn't just diverge in strategy. They diverged in who their customer is.

For firms already on the Microsoft stack with Claude embedded in enterprise tooling, this is validation. But the more interesting question isn't which platform won. It's what that win changes for the clients we serve. Platform selection used to feel like a technology decision you could defer. It's becoming a strategic one you can't afford to skip. Clients are asking which AI vendor to trust for multi-year production systems. Meanwhile, we are seeing such cost compression that signing anything multi-year feels like we may end up overpaying. The question they're really asking is: who can I build on that won't blow up my procurement process, my security review, my budget, or my board's risk appetite next year?

The answer is crystallizing. Firms that have a clear, defensible position on that question will close more work than the ones still saying "it depends." Platform-agnostic was a safe answer when the landscape was genuinely uncertain. That's less true now.