Microsoft Azure Opens to AI: Recomb's Analysis

Today (or rather, until today) software has been built according to the needs of its users but, above all, according to the needs or beliefs of those who program and create it!

?? ♂? Please raise your hand if you've never been frustrated by a programmer's decision or puzzled by a seemingly complex technical choice.

? No one?

?? Well, then let's move on.

?? Natural language input and output as a solution to many of the world's ills caused by programmers

With this cryptic phrase, we commented on the recent announcement by Microsoft to open Azure to artificial intelligence (AI), allowing companies to use their proprietary data and user queries with OpenAI's GPT-4 model within Azure.

What does all this mean?

?? In simpler terms, this means that companies storing their data in Microsoft's cloud can use AI to determine which data to use and how to formulate the answer.

?? The system works by taking a user's query, determining the necessary corporate data, and combining this into a new query. This new query is then processed by an OpenAI model within Azure, which generates an answer that is sent back to the user.

?? Microsoft guarantees that these artificial intelligence models are hosted and managed exclusively within its cloud environment. This means that OpenAI, the provider of the "matrix", has no access to the models themselves, nor to customer data, their queries or the results produced. Furthermore, this information is not used to train other services.

It is therefore important to underline that customers do not have access to the models or data of other customers. In practice, each Azure customer works with their own private and isolated version of the OpenAI GPT model!

?? This essentially means two things.

1?? Firstly, Microsoft's service is set to overcome a significant barrier that has hindered full AI adoption across all industries, not just modern companies. Regardless of how a company organizes or manages its data, the service will enable an AI to understand the request, select, and potentially cross-reference the necessary data to fulfill it.

2??Secondly, both the request and the machine's response will be in natural language! There's no need for complex queries to get sales data for a product in a specific period or segment. You just need to ask in natural language, and the AI will handle the rest, including providing a clear and understandable answer.

?No more software with complex interfaces. No more command chains to obtain a specific piece of data and, above all, no more it can't be done to a new implementation request!

??All data always available, always ready, extracted and enjoyed through language. All of this is damn exciting!

Got the intro's point now? ??

?? To learn more, read here: https://www.theregister.com/2023/06/22/microsoft_azure_ai_data/