An Azure service that provides access to OpenAI’s GPT-3 models with enterprise capabilities.
For Foundry models sold by Azure, the main place to review data handling details is the Data, privacy, and security for Models sold by Azure in Microsoft Foundry article.
That document states that prompts, completions, embeddings, and training data:
- are not available to other customers
- are not available to OpenAI or other providers of Models sold by Azure
- are not used by providers of Models sold by Azure to improve their models or services
- are not used to train any generative AI foundation models without permission or instruction
- are not used to improve Microsoft or third-party products or services without explicit permission or instruction
It also states that:
- fine-tuned models are available exclusively for the customer’s use
- Microsoft hosts these models in Microsoft’s Azure environment
- these models do not interact with services operated by the model providers
If the deployment type is Global Standard, the FAQ adds that:
- prompts and outputs are processed using Azure’s global infrastructure
- data stored at rest remains in the designated Azure geography
- data may be processed for inferencing in any Azure location
- prompts and outputs are not shared with the model provider
- customer data is not used to retrain the models
For a broader entry point, the Overview of Microsoft Foundry Models page links to the privacy article for models sold by Azure.
The provided material does not give a single general retention duration for all Foundry models sold by Azure. It only states that models sold by Azure store and process data to provide the service and to monitor for uses that violate applicable product terms.