Production AI Institute · Public record
<- AI Data Use Index
Microsoft Azure · AI development platform

Microsoft Foundry direct models

No foundation-model training without permission

Microsoft says prompts, outputs, embeddings, and training data for Foundry direct models are not used to train generative-AI foundation models without customer permission or instruction.

PSF standardPSF data protectionAll vendor disclosures
Training use

Customer inputs and outputs are not used to improve Microsoft or third-party products and are not used to train foundation models without explicit permission or instruction.

How to opt out

No opt-out is required for the stated training baseline. Deployment type, stateful features, feedback, and logging still require configuration review.

Private content

Microsoft says direct-model data is not available to other customers or model providers. Processing geography depends on regional, Data Zone, or Global deployment choices.

Retention

Base-model inference is described as stateless, while Responses, Assistants, stored completions, batch, files, and vector stores can persist or route data differently.

Human review

Abuse monitoring can select samples for automated and, when necessary, human review; qualifying managed customers may apply for modified monitoring.

Start here: production AI

Foundational reference pages for teams adopting, deploying, and operating production AI with inspectable controls.

What is production AI?AI agent production ready checklistAI adoption operating guideEnterprise AI deployment guideFive-year automation roadmapWorkflowOS open-source PSF studioPSF standard →
Public record

This record is maintained by PAI and free to cite. If something is wrong or missing, tell us. Corrections and source suggestions keep the record honest.

Records are free to cite. citation guidance.