I’m old enough now to have experienced several distinct waves of transformation brought by digital technology. As a kid, the personal computing revolution captured my imagination and energy with gaming and programming and new ways to create and do work. As a young adult, personal computers were everywhere and the internet and the World Wide Web connected them, and more importantly, the people using them, in ways that allowed communication and information to flow freely, and for work, commerce, creativity and leisure to be done in radically different ways. In my 30s, the smartphone and an incredible ecosystem of apps and services extended the internet to our pockets, making our connections to information and each other more ubiquitous, helping us navigate our way through the physical world, allowing us to buy almost any good or service we can think of, entertaining us in wonderful new ways, and making collaboration to get our work done more powerful than ever.
Even though that’s already a lot of transformation in a short period of time, and technology has never been more present in our lives, I feel like we’re just getting started. The next wave – one that’s already happening – comes when cheap connected devices with powerful sensors become truly ubiquitous in all of our physical environments, and when those devices become powerful enough to use the techniques of artificial intelligence (AI) to interact with their surroundings and the people in them. We call this combination of connected devices with powerful sensors and AI the Intelligent Edge. A year ago, I shared my belief that the Intelligent Edge would unfold as a platform over the next several years in ways that would surprise us by its breadth and diversity. And it already has.
The Intelligent Edge is proving to be the last mile in the convergence of the digital and physical worlds. –whether it’s a mixed-reality device like HoloLens providing a technician with a digital overlay of analytics, diagnostics and documentation for a piece of equipment they are servicing, or smart devices making the places where we live, work and shop more responsive and interactive, safer and more efficient. Intelligent Edge technologies are already making our homes smarter, improving the yields of our farms, monitoring the environment, helping us navigate our work more effectively, and improving our health and safety.
We’re in the middle of a revolution that is more than just smart speakers, security cameras and clever thermostats. Right now, we have in excess of 12 billion devices connected to the internet. It’s forecast that by the end of this calendar year, that number will rise to 20 billion. We anticipate that billions more of these devices are going to connect to the internet in the next few years. It’s a staggering thought. This Internet of Things (IoT) is already many times larger than the universe of personal computers and smartphones combined, and devices on it are becoming more powerful and more intelligent every day. With the advent of 5G, with its higher throughput, lower latency to the cloud, and higher device densities at the edge, we are likely to see the growth of the Intelligent Edge accelerate even further.
It probably comes as no surprise that I’ve been super stoked by each of the big technology platform waves that I’ve personally experienced, from PCs, to the internet, to smartphones. The Intelligent Edge is no different. I can’t keep from tinkering with these technologies, and I’m a bit embarrassed to admit that I’m using bits of the Intelligent Edge platform to build, of all things, an AI-powered siphon vacuum coffee machine. Instead of screens and buttons, my machine has a camera, a microphone, a speaker, a small digital brain and a connection to the cloud. When you focus your attention on it, it notices, and will ask “Would you like a cup of coffee?” When you respond “Yes,” it guides you through the brewing process with a short dialogue. And if you like, it will remember you and your preferences so that you can get your next cup of coffee more quickly.
My coffee machine probably won’t be commercially viable, and no one should mistake my weekend tinkering for a product that might one day show up in the Microsoft store. But one thing that’s become very clear to me as I build this machine is this: The Intelligent Edge parts of the device are neither especially hard nor expensive. I’m having a tougher time designing a safe steam boiler than I am with the AI! The hardware I’m using to run some of the local AI is cheap and readily available, and the software techniques I’m using to split the AI computations between the edge and the cloud are relatively straightforward. The Intelligent Edge and Intelligent Cloud platform that’s already out there for everyone to use is already quite capable. And even though to some, my coffee machine sounds like a crazy sci-fi project, making it a reality doesn’t feel as challenging as writing my first PC program, internet service or mobile app felt in the early days of those platforms.
What I’m most excited about with the Intelligent Edge is not what we’ve already done, nor even what I can imagine might be done with this new platform, but rather, what others will imagine and create as tens of millions of developers, entrepreneurs and innovative thinkers start building new products and businesses with this technology. Given the magnitude of growth ahead of us, and the fact that the platform is becoming more powerful every day, the opportunities for creators, entrepreneurs and businesses are huge. As with any successful platform, the true measure of the Intelligent Edge’s success will be in the breadth and diversity of the things built on top of it. There, I have infinite faith in the vision and ambition of others.
The IoT Signals Report (an annual research survey published by Microsoft) identifies key, industry-relevant trends in IoT. The survey, conducted by individual interviews with more than 3,000 IoT professionals based in Europe, Asia and North America, found that IoT is considered mainstream. Businesses are seeing tremendous value and opportunity in their ability to improve their bottom lines through IoT adoption. Right now, we’re seeing significant advancements in what I call a new world order with the demise of Moore’s law and the collapse of Dennard scaling. This means that compute is no longer becoming cheap at the exact same time that machine learning is becoming an insatiable consumer of compute power. But while this shift is impacting PCs, we will still see a few years where the power and compute capabilities of Intelligent Edge devices will continue to improve exponentially without much increase in cost.
IoT devices that are part of the Intelligent Edge provide businesses with invaluable insights on how to transform processes for operational efficiencies, such as improving the maintenance of vital of equipment before a costly shutdown and accelerating innovation while simultaneously improving safety, for example. As the IoT landscape continues to expand, we can bank on critical breakthroughs in areas that benefit humanity, such as healthcare, conservation, sustainability, accessibility and disaster recovery.