Sometime in 2025, Windows stopped feeling like an operating system and started feeling like a demo for AI. Open Notepad to jot something down and there it was, asking you to summarize it. Launch Edge and copilot politely waved from the sidebar. Even apps like Microsoft Paint felt different, not because they got easier, but because they suddenly wanted to create, edit, and enhance images for you.
Microsoft hasn’t just added AI, they’ve integrated it into every aspect of the experience. And for a while that felt exciting. Then it started to feel…a bit much.
Microslop: The most popular roast on the internet
It was around this point that the Internet did what it does best. It coined a name: Microslop. Rough, catchy and brutally effective. Adapted from the broader term “AI slop,” which refers to low-quality, mass-produced AI outputs, the term quickly became shorthand for something more specific.
Not just bad AI, but also unwanted AI.
The kind that shows up uninvited, sits too close, and insists on helping when all you really wanted to do was write a shopping list. It reflected growing frustration with Microsoft’s software becoming louder, heavier and somewhat more unpredictable.
The backlash became so loud that even CEO Satya Nadella publicly rejected the idea of dismissing AI as “sloppy.” Ironically, this only accelerated the spread of the term. By early 2026, it had become a full-fledged cultural shorthand for dissatisfaction with Microsoft’s AI push and was even banned in some official communities. At this point, this was no longer just a meme. It was feedback.
The moment Microsoft blinked
For a while it felt like Microsoft would just move on. But then, in March 2026, in a surprisingly candid blog post titled “Our commitment to Windows quality“Microsoft confirmed what users had been saying for months. The company talked about improving reliability, reducing friction, and making Windows smoother and more reliable again. Among other things, Microsoft said it would also reduce Copilot’s presence in Windows.
And these weren’t just empty promises. Across multiple apps, the company reduced the number of entry points where AI appeared. Previously announced features, like deeper Copilot integrations into notifications, have been quietly shelved. Additionally, apps like Notepad, Photos, and Snipping Tool no longer have visible Copilot hooks.
On paper, it seems to be exactly what users were asking for. Less AI clutter. More focus. Of course the narrative became simple. Microsoft heard the backlash and scaled back measures. But like most simple narratives, this one doesn’t quite hold up.
Why Microsoft can’t just “turn off” AI
Here’s the thing. Microsoft can’t actually escape AI, even if it wants to. This is not a feature change. It is the foundation for everything the company is currently building. From the Azure infrastructure to Microsoft 365 to Windows itself, AI is firmly anchored in the strategy. Billions have already been invested. Entire product lines are then realigned.
Microsoft was one of the early backers (read: billions of dollars) of OpenAI, heavily integrating ChatGPT into its products, and then borrowing Claude AI from rival Anthropic to power Copilot — all while developing its own AI models. The AI push even spawned a whole new generation of laptops with Copilot+ branding and a dedicated Copilot key on the keyboard.
Yes, “absurd,” you might say.
Even now, as Microsoft reduces visible integrations, it’s still pushing Copilot into enterprise tools, workflows, and services. So what you see is not a retreat. It’s a recalibration. AI is not going away. It is simply repositioned by making it less visible but silently penetrating the foundations.
Stealth mode activated?
You can see this most clearly in the small details. Take Notepad for example. A year ago there was a bright Copilot button right in the interface. It was obvious, almost eager. In newer builds this button is no longer present. In its place is a much more neutral “writing tools” symbol. The features are still there. Rewrite, summarize, adjust tone. But the branding is gone. The volume is gone.
And this is not an isolated case. On Windows, Microsoft is reducing the frequency with which Copilot appears as a named feature while retaining the underlying functionality, from AI capabilities to advanced features and so on. Some call this “stealth slop.” AI that hasn’t disappeared, but has learned to avoid you. Fewer announcements, more availability.
What’s fascinating is that Microsoft’s core belief hasn’t changed at all. The company still sees AI as the future of computing. If anything, it doubles behind the scenes. What has changed is delivery. The first phase was about visibility. Send AI anywhere. Make sure users see it, notice it, and ultimately try it. That worked, but it also backfired.
People didn’t just notice AI. They felt overwhelmed by it.
Now we are in phase two. Integration. Microsoft is being more selective about where AI appears and how it behaves. Executives have even said they want to focus on AI experiences that are “truly useful” and not just generally available. It is a shift from proving ability to proving value.
The real change
Microsoft hasn’t actually “fixed” the problem, but that may not even be the right way to look at it. The backlash wasn’t about AI being bad; The point was that it was everywhere in a way that felt unnecessary and intrusive. This distinction is important. Even now, criticism of forced integrations and limited user control hasn’t completely disappeared, but at the same time Microsoft is clearly trying to clean things up with a more focused, less cluttered Windows experience.
What’s really changing is not the presence of AI, but how it feels. Instead of being a loud, intrusive feature, AI is being redesigned into something quieter and more natural. The goal now seems simple. Make it helpful without making it obvious. Because for AI to actually work on a large scale, it must not feel like an add-on. It must feel like it was always meant to be there.
That’s the lesson Microsoft apparently learned the hard way. AI has not been removed from Windows. It just made it less noticeable. Microsoft is no slouch in the AI space. Earlier this month, Microsoft announced not one, but three core AI models. The Phi series of open source models for small languages is quite popular and powerful.
By next year, Microsoft wants to release its own frontier models to compete with ChatGPT, Claude and Gemini. “We have to push the absolute limits,” Mustafa Suleyman, head of Microsoft’s AI efforts, said in an interview. As I said, the AI push will continue. I just hope it continues to evolve without tainting everything Microsoft offers to hundreds of millions of users around the world – including lifelong die-hard users like me!
