What Satya Nadella’s Top 5 Copilot Features at #MSBuild Reveal about the Future of AI Development
Microsoft Build (#msbuild) is a conference that highlights the latest innovations and trends in software development. This year, the conference focused on how developers can use Microsoft's AI (Artificial Intelligence) capabilities to create smarter and more efficient solutions.?
Microsoft branded its AI platform #Copilot, which integrates AI features into Microsoft products including Windows, GitHub, M365 and Dynamics. By embedding AI into these products, Microsoft aims to enable everyone and organization to achieve more with less effort.?
Among the 50+ new announcements made at Microsoft Build this year, Microsoft CEO Satya Nadella highlighted his top five favorite features in his keynote speech. These features are:?
Microsoft is providing Bing to ChatGPT to ground the answers. The process of Retrieving Augmented Generation (RAG) additional information outside the model to ground the answers provided back.
The goal of this process is to ensure that large language models have up to date and relevant information that is being asked about outside of the training data set.
This new feature's impact will allow ChatGPT to be able to more accurately and consistently answer questions outside of its training model.?
The most impactful announcement at Microsoft build was the integration of Copilot into windows. The ability of any user to engage and communicate with complex app specific functionality through an API simply through a chat window is going to change how people leverage computers and how they view applications.?
?“This is going to make every user a power user of Windows.” – Satya Nadella
During the demo Satya highlighted how somebody looking to build a logo for a coffee shop was simply prompted to use an already installed app of Adobe Express to generate a logo. This is one example of how to use the ability of tapping into the robust functionality that exists within installed applications.?
Even complex applications that are installed on the windows platform will now be able to tap into Windows Copilot through an API.?
Microsoft announced this week the Copilot stack. Microsoft at its core has always been a platform company and this has never been more seen than its announcement of the Copilot stack. This Copilot stack will allow any developer to create a plug in for Copilot that is available anywhere Copilot is available.?
?
Microsoft touted the fact that they have been able to move fast over the last few months because of the copilot stack and being able to have a solid foundation on which all developers internally at Microsoft can build from.?
领英推荐
?
During the demo we saw Bing chat utilize the context of a webpage to pull the ingredients list and then leverage Instacart to put them into a shopping cart. The extensibility to create the plugin in one place and have it be accessible via multiple locations increases the value proposition for developers when building a plugin on ChatGPT.?
?
By providing a framework for developers around the world to engage with ChatGPT and Microsoft's Copilot, Microsoft looks to make Copilot a foundational platform on which users can interact with various companies and perform complex tasks in a single pane of glass.?
?
Microsoft has released information on GitHub for those interested in build a plug-in on the OpenAI.?
Microsoft has continued to make available all their best practices through their Azure services. By being able to build anywhere along the stack with one consistent AI studio will allow developers to utilize Microsoft best practices including the investments Microsoft has made into AI safety.??
Microsoft continues to allow developers to bring in not just the open AI models, but developers can also create their own models as well as use other existing models as part of their AI capabilities as well. Bringing simplicity to complex items is what Microsoft has focused on.?
One of the hardest challenges for organizations is master data management. Having one source of truth when it comes to data that can be leveraged for machine learning visualizing data as well as ad hoc analysis, is what Microsoft Fabric seeks to achieve. Organizational data is stored in Microsoft Fabric OneLake, which is designed to be a single place for all analytic data.??
?
By storing the data in a single location governance and security can be applied at the base level of OneLake. Administrators can set security of the data at the OneLake level thereby allowing users throughout the organization to use the data without worrying about security.?
?
The integration of OneLake into Microsoft's other products including Azure machine learning as well as M365 seeks to empower every user whether a data scientist or a business user to access data from across the organization securely.??