What Can Azure OpenAI Do for You?
Microsoft MVP Eric D. Boyd presents a sneak peek of his anticipated Live! 360 Azure OpenAI discussion.
Azure OpenAI Service symbolizes the collaborative effort between Microsoft and OpenAI to provide customers with sophisticated language AI functionalities. This service hosts a range of high-fidelity models including GPT-4, GPT-3, Codex, DALL-E, and Whisper, all safeguarded by hardened security and enterprise-level compliance. And, with Microsoft's "all in" approach to generative AI, its capabilities are growing by the day.
Azure pro and Microsoft MVP Eric D. Boyd will be breaking down what you need to know about OpenAI's integration in Microsoft Azure -- from what's new, what you need to know to get started and how Microsoft is ensuring your enterprise data will be safe with the growth of AI – at his popular upcoming Live! 360 session, titled "Fast Focus: Inside Azure OpenAI Service."
Ahead of that, Boyd sits down with Redmond to discuss how Azure OpenAI is changing businesses, what Microsoft is doing to promote responsible AI use and why you won't want to miss his full talk at Live! 360 in Orlando, Fla. this November.
Redmond: OpenAI technologies like GPT and ChatGPT can enable huge productivity gains. Can you give us a sneak peek into how Azure OpenAI makes this easier for businesses?
Boyd: Azure OpenAI delivers the OpenAI GPT models on Azure's robust hyperscale cloud. Azure provides additional responsible AI capabilities including safety layers and content moderation on the inputs and outputs of the models. Azure OpenAI is integrated with Microsoft Entra ID for consistent identity and access management. Azure OpenAI also provides networking capability to integrate the OpenAI service into your private Azure Virtual Network.
For those who are new to AI or OpenAI's offerings, what would you say is the most critical thing to understand before diving into Azure OpenAI?
I think it's important to understand what OpenAI's models do and what they don't do. They are simply large language models that are trained with lots of words, or more accurately tokens, which are word fragments. They are designed to suggest the next set of words or tokens that make sense given some input and guidance. They don't think or have any sort of human like emotions or consciousness, they are simply suggesting tokens based on patterns of tokens they have seen before.
You're promising to share some specific use cases for integrating OpenAI into various solutions. Could you provide a brief example that attendees will get to explore in detail during the presentation?
Whenever you have a need to create natural language text, for different scenarios, over and over again, you should consider whether OpenAI models are a good fit.
For example, if you are interacting in chat and conversation with an automated conversational chat bot and have question and answering scenarios, OpenAI's GPT models are great at this. If you are writing content like articles, blog posts, marketing content or even just rewriting and editing the structure of the content you've written, that's another great use case for OpenAI. If you want to analyze and reason over existing text and content, or summarize content, OpenAI is great at this.
Microsoft's commitment to responsible AI is a hot topic. How does this manifest in Azure OpenAI's service?
Microsoft is committed to responsible AI across six principles: fairness, reliability and safety, privacy and security, inclusiveness, transparency and accountability. And we see these principles manifested in the Azure OpenAI service. A couple of important things to point out is with Azure OpenAI your data is encrypted at rest, it isn't used to retrain the models, and it isn't ever sent to OpenAI. And those are some important factors for customers when using AI and cloud services in general.
The event is labeled as "Introductory to Intermediate." Could you elaborate on what types of skills or knowledge attendees should ideally have to get the most out of this session?
If you have heard about ChatGPT and OpenAI and are interested in what's going on in the world of AI and large language models, but haven't had time to read up and explore how you can take advantage of these capabilities in your organizations, this session is a great way to get some foundational knowledge of what's possible.
Azure has many services and OpenAI is a popular platform. What sets Azure OpenAI apart, and why should someone not miss your session on this topic?
Azure has many services, and even many AI services, including the Azure AI services that used to be known as Cognitive Services, as well as Azure Machine Learning, and the foundational Azure AI infrastructure that makes it all possible. And OpenAI exists outside of Azure. But what sets Azure OpenAI apart from the many services in Azure is the ability to generate high-quality natural language and even code, in scenarios like chat bots, content authoring, reasoning, and even summarization. And the power of the Azure platform with the global data center footprint, compliance certifications, and the integration of Azure OpenAI with other Microsoft and Azure services like Entra ID and Azure Virtual Networks, sets Azure OpenAI apart.