Within the last 50 years we had several technological booms that completely, and irreversibly changed our lives. Starting with the personal computer boom happening around 1980’s, then the Internet boom in the 1990s and 2000’s and Mobile Revolution in the 2000’s. For the past two years (as of writing this blog in 2024) we have entered yet another revolution – the age of Artificial Intelligence. AI is perhaps the biggest and most important technological advancement in more than 40 years according to Microsoft founder Bill Gates.
In this blog I would like to introduce you to the AI offering from Microsoft. From productivity enhancing Microsoft Copilot, to empowering developers with Azure AI and GitHub Copilot.
Before 2020
Microsoft’s AI work can be split into two phases: what came before 2020 and what came after. Let’s first take a look at what Microsoft developed before 2020. Microsoft began its AI journey with the launch of Microsoft Cognitive Services in 2015, a collection of APIs that enabled developers to add AI capabilities such as vision, speech, language, and decision-making to their applications. Following that, Azure Machine Learning was introduced, a cloud-based service for building, training, and deploying machine learning models. In 2016, Microsoft released the Bot Framework, a platform for building and connecting intelligent bots that could interact with users across various channels. Finally, in 2017, we saw the introduction of Azure AI, a suite of AI services and tools on the Azure cloud platform, including Azure Cognitive Services, Azure Machine Learning, and Azure Bot Services.
After 2020
In 2020, we were introduced to OpenAI’s GPT-3, which quite literally changed how many of us work forever. OpenAI’s groundbreaking work on large language models (LLMs) paved the way for what Microsoft would later offer in their AI lineup. In 2021, we saw the launch of GitHub Copilot, an AI-powered code completion tool that helps developers write code faster and more efficiently. That same year, Microsoft also introduced the Azure OpenAI Service, a platform providing access to OpenAI’s powerful language models, including GPT-3, through Azure.
Microsoft Copilot
Imagine having a personal assistant that not only understands your tasks but also anticipates your needs, offers insightful suggestions, and automates your routine processes—without being that over-eager friend who tries to help but ends up causing chaos! Microsoft Copilot is exactly that.
Before Copilot became what we know today, there was Bing, an AI-enhanced chat similar to GPT, which Bing used as its engine. Later, Bing was renamed as Copilot, and it expanded to seamlessly integrate with all Microsoft 365 apps like Word, Excel, PowerPoint, and Outlook. It helps us draft documents, create presentations, analyze data, merge emails for greater efficiency, and write all kinds of texts—just like this one.
If I were to list Copilot’s biggest strengths, these two would top the list:
Copilot’s Intelligence
With the power of AI, Copilot can easily understand and respond to us in natural language. It’s like talking to an all-knowing friend or professor who can explain the most complex quantum physics theories in a way a 5-year-old could understand.
Copilot’s Contextual Understanding
Another great feature of Copilot is its ability to understand the context of the work you’re doing. It can pull relevant information from your documents, emails, and other sources to provide accurate and helpful responses.
Phi-3 Models
We’ll return to Copilot in a bit with another exciting technology, but for now, let’s explore another offering from Microsoft introduced in 2023—the Phi-3 Models.
Phi-3 Models are a family of small language models (SMLs) designed to deliver high performance at a lower cost with reduced computational demands. These models can be deployed in various environments, including the cloud, at the edge, or directly on devices. This flexibility makes them ideal for real-time interactions, autonomous systems, and other scenarios. Currently, there are four different models:
Phi-3-mini
Starting with the smallest of the models, Phi-3- is ideal for applications requiring high performance with lower computational resources.
Phi-3-small
The middle-child of Phi-family is Phi-3-small, which is suitable for more complex tasks that need a balance between performance and resource usage.
Phi-3-medium
We then have the “big kid”, Phi-3-medium. The medium model of Phi-family is best for applications demanding high accuracy and relevance, such as detailed data analysis and advanced natural language processing.
Phi-3-Vision
Then there is the vision model, a state-of-the-art multimodal model designed to handle both text and visual inputs.
Microsoft Fabric
This year (2024), Microsoft introduced us to Microsoft Fabric. Fabric is a comprehensive, AI-powered data platform designed to unify and streamline data management, analytics, and business intelligence. It integrates various data services into a single platform, including data movement, processing, ingestion, transformation, and real-time event routing. These services help reduce the complexity and costs of managing multiple data systems. Microsoft Fabric uses a single data lake—OneLake—allowing everyone on your team to connect to the same data from anywhere and work from the same copy across different engines.
Copilot+ PC
As I mentioned earlier, let’s circle back to Copilot for the final point on this list—this time with Copilot+ PC. Copilot+ PC is the latest release from Microsoft (as of writing this blog). It integrates Copilot into the operating system and various Microsoft 365 applications. With PC’s running Snapdragon X Series processor, capable of performing over 40 trillion operations per second, we can expect the fastest and most secure Windows PC ever built—at least according to Microsoft. But what functionality can you expect from such PC? Here is a list of what can be expected:
- Real-Time translations
- Image Generation
- Co-Creator Tools
- Security Enhancements
- Voice Assistance
- And the newest feature – Recall
Recall
One of the most interesting features of Copilot+ PCs is Recall. It uses Windows Copilot Runtime to help you find anything you’ve seen on your PC by essentially taking snapshots of your screen. How can this benefit you? Imagine you want to continue working on that video of a banana you were editing in DaVinci Resolve some long time ago. With Recall, you can browse through all the snapshots and jump back to the exact moment you left the editing software, allowing you to pick up right where you left off. These “recalls” are stored locally on your computer and aren’t uploaded to the cloud, ensuring your data stays secure.
Conclusion
If you ask me, we’ve entered an incredibly exciting time for technology. With all the AI research happening around the world, and with OpenAI, backed by Microsoft, leading the way, I’m sure we can expect some amazing innovations that will revolutionize our daily lives even more than ChatGPT and Copilot already have. But in order to get most out of that amazing technology, information management, security and protection is more important than ever before. Get in touch with us at contact@infotechtion.com and we will help you prepare your organization for the age of of AI!
