Explore Azure OpenAI

  • 5 minutes

The Azure OpenAI Service offers API access to advanced language models such as GPT-4, GPT-4 Turbo with Vision, GPT-3.5-Turbo, and the Embeddings series. These models are flexible, supporting tasks like content generation, summarization, and image understanding. You can access the models through REST APIs, language specific SDKs, or directly via the Azure AI Foundry portal.

Security, access control, and ethical AI use

Generative AI models available through Azure OpenAI provide powerful tools for enhancing applications with complex language and image processing functions, all within the robust and secure Azure environment. Azure OpenAI models include security features like virtual network support and managed identities ensure safe data handling, making model management secure, and straightforward.

Access to Azure OpenAI is behind a limited access policy to ensure adherence to Microsoft's standards for ethical AI. As part of the limited access policy agreement, teams or individuals tasked with developing and deploying AI systems should work to identify, measure, and mitigate harm. Users needing to apply for service access can do so at https://aka.ms/oai/access.

Note

Azure OpenAI Service has been released with limited access to support the ethical use of the service. You can read Microsoft's Transparency note for Azure OpenAI Service here.

Familiarizing yourself with the available features for submitting prompts and generating responses allows you to effectively implement Azure OpenAI to enhance your applications, ensuring they're both powerful and in line with ethical AI practices.

Once you create an Azure OpenAI resource and deploy the desired models within the Azure ecosystem, you or your businesses can use the advanced AI technologies while keeping their operations scalable, secure, and compliant. One of the model families called embeddings is designed to enhance your vector search solution, enabling a Retrieval Augmentation Generation (RAG) systems.

RAG enhances the system's understanding and creation of text by providing improved contextual information to the generative AI model with your own data. Azure OpenAI combined with your own data in vCore-based Azure Cosmos DB for MongoDB produces more accurate and contextually relevant responses.

Using the Azure OpenAI service in your applications allows you to tap into advanced AI features, boosting efficiency and flexibility in your business operations all while adhering to principles of ethical AI usage.