OpenAI Private Intelligence: tighter data controls for businesses

OpenAI launched Private Intelligence, a set of controls for companies. It includes zero data retention with 'private safety processing', where anything flagged is encrypted into storage the customer owns and deleted after 30 days; Private Inference, in preview, which runs requests on sealed 'confidential computing' hardware; and controls over what agents may do and approve. It covers 'eligible workflows' only, and there are no prices, no model list, no date for Private Inference and no named customers yet.

Why it mattersIf a client has refused ChatGPT or the API over data concerns, this is the conversation to have with OpenAI's sales team. It's for enterprise accounts for now, so ask exactly which workflows qualify.

Curated by Sara Simeone as part of our human editorial coverage. The original sources are linked below; curation is distinct from a Hype or Fact? assessment.

How we knowCompany figures are labelled as their own claims; anything we couldn't verify is labelled as a report or a claim.