What products and services will benefit most from the latest advances in artificial intelligence? Everyone is following the issue closely and there is curiosity as to whether new companies will be able to better take advantage of AI improvements or if cloud platforms are the best positioned. Or perhaps it will be companies that build generative AI models themselves.
Despite all the considerations, a key type of software is missing from the discussion: the operating system.
Recently, Microsoft announced a number of new products and features, including the Microsoft copilot for Windows 11. Calling it an “everyday AI companion” that will be included in Windows and other Microsoft products, the company says the product is an extension of previous efforts to bring Bing's first AI tools to Windows. It is the first update to an operating system that causes excitement in a long time.
The question may be asked as to how much the computing experience has really changed from Windows 10 to Windows 11, or from the last version of macOS used to whatever any user is using now. The same question works for iOS and other everyday operating systems. The answer is similar for everyone: not much. Unless the user uses some variant of Linux, of course.
This is because operating systems have become docile computing layers that mostly serve as a base for running other applications. It doesn't matter much which operating system is used, because we use the same web browser, for example, on all of them. You don't really need to know much about the operating system the browser is running at that particular time; You only need to use web applications.
This happy stagnation is not bad, of course. The fact that iOS still shows you a grid of apps when you launch it is because it is a simple and intuitive way to show the user their set of apps. Microsoft is similarly investing in the Start menu, which has worked quite well for decades. And different versions of Android have their own versions of the application menu.
For the apps we use today, our mobile and desktop operating systems feel solid, uncluttered, and, compared to previous periods, quite secure. And they are mostly free if the hardware is new. It's a winning formula for consumers who get a lot of bang for their buck.
Similar to how modern smartphones seem to have been "figured out" and are now only gradually being improved, for some time it has felt like major operating system companies have been working to polish and perfect, rather than reinvent. Since consumers seem quite satisfied, so far everything has been going well.
New AI technology could disrupt the status quo, and Microsoft's addition of AI at the operating system level could be a big step in that direction.
The voice assistant trap
When Microsoft began to move away from Cortana, it caused some discouragement. Apple's Siri and Google's own suite of voice assistants had blind spots, so if a potential competitor fell behind it meant there would be less pressure on rival products to improve. Given many users' chronic use of iOS and their penchant for sending lots of slightly garbled Siri-transcribed text messages, it's long been hoped that we could get more done by talking to devices.
The question now is how good Microsoft Copilot will be and what it will be like to use it every day.
It's not that working on Windows is very different from working on Mac OS; It's just that we tend to use different types of equipment for professional and personal issues. With this news, an interesting curiosity is generated about the impact of AI on the personal aspect.
The prospect of having an AI-powered generative digital assistant that can execute tasks for me at the operating system level is incredibly exciting.
One problem with using AI within a particular app or service is that its intelligence is often limited to the data the app in question has about me or my company, and the scope of the program itself. A CRM may have some really cool AI capabilities, but that won't help outside of that realm. When we take AI to the operating system layer, it has the potential to encompass many, if not all, other applications and services, creating the possibility of something more powerful and broader.
Only an intelligent operating system can chain complex tasks between applications from different companies; I think that nothing else will have the necessary permissions. And now that we're getting closer to that reality (hopefully all OS vendors will follow similar paths as Microsoft), we're moving away from AI as a point solution and toward AI for everyday use.
And seeing as generative AI technology improves rapidly, we may be entering an era where operating systems feel and perform differently over time, and their updates could bring more than just security patches. Interesting, right?
Yes, but there are also some disadvantages to consider.
How will we pay for it?
Do we remember Windows Vista? The version we use? Was it Windows Vista Starter, Home Premium, Home Basic or Ultimate? Did the company use Vista Business or Vista Enterprise? I suppose most had one of the Home versions, but, it's a blind shot.
All of those different versions of Vista offered different features and sold at different prices. A lot of money could have been spent on Vista: apparently up to $399,99 for the Ultimate version. Can we imagine paying that amount today for an operating system upgrade? Hardly. Unless we are managing a company, of course.
Since most of us purchase an operating system when we buy a new computer, we don't think much about its retail value. What's more, we don't think about how good the included cost really is. For example, when buying a computer with Windows 11, from the first moment it is at your discretion. You never think about the current version, the need to update it or the cost.
At least two of those things are going to change in the AI-OS era. The current version of the operating system will be more important, as it can be anticipated that the AI within the operating system layer will improve rapidly, meaning you will want to update more frequently. And that may come at a cost.
It's not cheap to run generative AI models, and while those costs will decrease over time, if you're going to interact dozens of times a day with Windows Copilot, as Microsoft seems to expect, the world will generate billions of generative AI calls every day. Maybe some of that computer work will be pushed to the limit, but I doubt it will all be like that. That would generate a huge core computing cost associated with generative AI digital assistants at the operating system level, and that cost would have to be paid for somehow.
Perhaps that's why Microsoft is working to generate more ads on Windows; You will need a recurring revenue stream from Windows users to pay your AI costs.
The company could also limit Microsoft Copilot to Microsoft 365 subscribers, but that would mean most people won't be able to use it. (There were 67 million Microsoft 365 consumer subscribers, according to the company's latest earnings report; In contrast, there are around 1.400 billion monthly Windows 10 and 11 users). No, Copilot will have to be incredibly cheap if it is going to be used by the majority of people. This could require even more territory in an operating system for ads, which it's easy to assume will be hated by everyone.
Even so, the willingness of certain users like mine to pay for digital services is higher than average, so I don't want my personal biases to cloud my view of the market. How companies answer this question about costs could divide the operating systems market in an interesting and competitive way.
So even the bad news isn't that bad. And the good news is that while AI is interesting when used within applications, we see the first steps in bringing something truly new and powerful to the operating system layer. It was time: to be able to talk to our computers to do more things faster.
