In early May, OpenAI was raising funds for a venture focused on enterprise deployment, with $4 billion raised from 19 investors at a $10 billion valuation. Hours later, a competitor announced an equivalent venture for enterprise AI services.
When two rivals reach the same conclusion on the same day, the diagnosis is usually sound. Here it concerns distribution: selling access through an API reaches people who already know how to build, and most corporate money sits with people who don't.
Large companies don't buy models, they buy outcomes inside a process that already exists, with legacy systems, undocumented exceptions and people who have to keep working during the change.
Structuring that as a separate venture rather than an internal department has financial logic. Deployment services carry lower margins and slower growth than selling access, and mixing the two in one structure worsens the read on both.
For technology service providers, the signal is favourable and worth recording: the model makers themselves acknowledged the deployment layer is a separate business, with its own skills and margins.