Microsoft AI ambitions struggle to meet goals

winjer

Member

Microsoft faces reality check on AI ambitions as Copilot and Foundry struggle to meet goals


It's hard to argue that we aren't in the midst of an AI bubble – analysts and AI CEOs largely agree. While there are no signs of a crash yet, Microsoft may be the first company to show signs that it overestimated enterprise and consumer interest.
Microsoft's push to make artificial intelligence the core of its product strategy is running into resistance from the very customers it expected to embrace it. Multiple Azure sales units fell short of ambitious growth targets tied to Foundry, Microsoft's marketplace for AI models and agent-building tools.

In response, the company reduced those growth expectations for the current fiscal year, an unusual step for a firm that typically raises quotas annually. The shortfall suggests enterprises remain hesitant to spend more on AI, even as Microsoft promotes agentic systems as the next major shift in workplace computing.

Foundry is central to Microsoft's vision for autonomous AI agents capable of handling multistep tasks with minimal oversight. The company pitched these tools as a way for businesses to automate everything from data processing to report writing, and reinforced the message with new Word, Excel, and PowerPoint agents revealed at recent developer events.
However, many companies that invested in these tools are not using them, reflecting persistent concerns about accuracy, brittleness, and the risk of high-stakes mistakes. Anonymous sources told The Information that one Azure unit asked salespeople to grow Foundry spending by 50 percent last fiscal year, but fewer than one-fifth met that goal. Seemingly in response, Microsoft lowered the target to roughly 25 percent for the current year.

2025-12-10-image-16-j_1100.webp

For a company that has spent years embedding AI into its operating system, office suite, and cloud tools, the pattern points to a broader disconnect: customers reject new technologies when a company forces them into existing products as default features rather than optional enhancements.

Microsoft went on the offensive after shares tumbled by over three percent when The Information broke the news. "The Information's story inaccurately combines the concepts of growth and sales quotas," a Microsoft spokesperson told Bloomberg. "Aggregate sales quotas for AI products have not been lowered."

Analysts at Jefferies brokerage firm supported Microsoft's position, but its stock still dropped sharply as investors weighed the prospect of softer-than-expected enterprise demand.

Microsoft's predicament highlights the tension across an industry that has "irrationally" invested billions in AI infrastructure, betting that widespread adoption is imminent. Businesses remain cautious, and consumers remain skeptical. Microsoft's aggressive AI-everywhere strategy now serves as an example of the current disconnect that has contributed to the largest economic bubble since the dotcom.

Analysts warn that the AI market will eventually self-correct, creating clear winners and losers. If Microsoft does not adopt a more consumer-friendly approach, it risks losing billions in AI investment.

Nick Offerman Smile GIF


Let us hope this is the first sign of the AI bubble popping, so we can go back to normal prices for hardware. And for all this push for agentic crap to be dialed down.
 
Last edited:
Well no shit. AI should be use as a tool to help, not to substitute the human factor. They dont know what generic means. Copilot make it worse.
 
Last edited:
Chatbots are just that: chatbots. Useless without supervision. They are only as good as the human user. How these billion / trillion dollar companies do not understand that.
 
They are pushing it like crazy for the Enterprise and even though it's a partially captive audience, it's not quite there considering the costs.

They are rolling in more "free" Copilot functionality into Copilot Chat (comes with business 365 sub) as the separate $30-40 license fee or separate Security Copilot fee are pretty high and so are a lot of Foundry costs. Foundry, to be fair, has a lot of governance and controls around agents and agentic actions which you really don't get if you go with mishmash of products.

It's going to be interesting to see what happens next year.
 
Last edited:
The problem with Microsoft is their enterprise users are power users who have spent decades building fluency, muscle memory, and shortcuts to do their jobs. Introducing AI agents into a workflow that is already on razor tight deadlines and margins is going to face resistance.

Now the problem for Microsoft and OpenAI is that Google - for the first time since the early Android days - is actually putting effort into something, and with the amount of talented engineers they have, they will probably stay in the lead a bit longer. It also helps that unlike Microsoft and OpenAI, google is experienced in making consumer products.

Microsoft has a big hill to climb.

Chatbots are just that: chatbots. Useless without supervision. They are only as good as the human user. How these billion / trillion dollar companies do not understand that.
That was always going to be step 2. Actual "agents" that you can assign tasks to are being tested and should be the next evolution.
 
I think most users will recoil at the thought of having an agentic OS. An AI rummaging through your private shit? No thanks.
Still, don't expect a crash, it'll find its uses. Animation, movie special effects. If you work in those fields, don't be John Henry, be the guy operating the steam-powered drill.

 
All the corporations were so busy trying to create an AI only future they forgot to ask their customers if they actually want their product or services.
 
I was so god damn tired of windows 11, the outlook client, and copilot being pushed everywhere into my system no matter what i did. So i bought a mac mini and it just works, no forced updates that breaks half my system. My self built pc just sits and collects dusts until a game comes out i want to play.
 
Last edited:
All the corporations were so busy trying to create an AI only future they forgot to ask their customers if they actually want their product or services.

What this really showed is how much the leaders of these companies want to get rid of human workers.
They are sinking billions on tools that can replace people, so they can fire workers. And they don't even know how well it's going to work out.
But this desire is so strong with these sociopaths, that they are willing to risk billions to achieve it.
 
Last edited:
The ai bubble is all about investor interest. These companies are not buying up ram to support ai as much as it's like saying " see we are buying all this ram for ai so buy our stock! ". If investor interest gets too low then data center companies like ms will stop with its bs.

The bad part is once prices are raised they very rarely ever go back to as low as before.
 
BTW, there is a new tool to curb stomp all that AI stuff in Windows 11.

RemoveWindowsAI script promises to strip Windows 11 of Copilot, Recall, and other AI features



RemoveWindowsAI is a PowerShell script on GitHub that aims to strip Windows 11 of built in AI features. It is made by a developer who goes by zoicware, and it is designed to run through a single command or a simple GUI wrapper.

The script exists because Windows 11 keeps adding AI features with each release, and 25H2 is the update people keep pointing at. A lot of Windows users are tired of constant AI features they did not ask for. RemoveWindowsAI is one answer to that frustration. It targets AI that shows up in the UI, and AI that sits deeper in packages and services.

Here is what it does, based on its own feature list. It can flip registry and policy switches to disable Copilot, Recall, and Input Insights typing data collection. It also targets Copilot inside Microsoft Edge, plus AI parts in apps like Paint (including Image Creator), Notepad (rewrite feature), Voice Access (AI voice effects), and AI in Settings search. It removes the AI Fabric Service and disables AI Actions. It can hide the "AI Components" settings page. It can remove Recall's optional feature and forcibly remove Recall scheduled tasks.
The heavier parts are where this turns into a nuclear option. The script can remove AI Appx packages, including some labeled nonremovable, and it can also try to remove locked AI packages from the Windows CBS servicing store. It also claims it can install a custom Windows Update package meant to prevent AI packages from being reinstalled later. The project also includes a manual disablement guide, since not everything can be handled automatically.

We do not recommend tools like this for most users. If you are mainly tired of seeing Copilot or Copilot+ prompts, start with the simple route, uninstall the Copilot app where Windows allows it, or disable it through standard policy or registry controls. If you still want to experiment, install Windows in a free virtual machine tool like VirtualBox or Hyper-V first, and test the script there before you touch your main install.

Script Features

  • Disable Registry Keys
    • Disable Copilot
    • Disable Recall
    • Disable Input Insights and typing data harvesting
    • Copilot in Edge
    • Image Creator in Paint
    • Remove AI Fabric Service
    • Disable AI Actions
    • Disable AI in Paint
    • Disable Voice Access
    • Disable AI Voice Effects
    • Disable AI in Settings Search
  • Prevent Reinstall of AI Packages
    • Installs custom Windows Update package to prevent reinstall of AI packages in the CBS (Component-Based Servicing) store
  • Disable Copilot policies
    • Disables policies related to Copilot and Recall in IntegratedServicesRegionPolicySet.json
  • Remove AI Appx Packages
    • Removes all AI appx packages including Nonremovable packages and WindowsWorkload
  • Remove Recall Optional Feature
  • Remove AI Packages in CBS
    • This will remove hidden and locked AI packages in the CBS (Component-Based Servicing) store
  • Remove AI Files
    • This will do a full system cleanup removing all remaining AI installers, registry keys, and package files
  • Hide AI Components
    • This will hide the settings page AI Components
  • Disable Rewrite AI Feature in Notepad
  • Remove Recall Tasks
    • Forceably removes all instances of Recall's scheduled tasks
 
I need AI to raise my future children according to Sam Altman.
I'm kinda starting to wonder if its not too far fetched of an idea that the end goal of these big tech companies has become to utterly weaken people, mentally and physically, to make them hopelessly dependent on their "gracious" technology.

Tech used to be fun and genuinely served a purpose to aide and push society forward. Now its starting to look like that intent has been overturned.
 
What this really showed is how much the leaders of these companies want to get rid of human workers.
They are sinking billions on tools that can replace people, so they can fire workers. And they don't even know how well it's going to work.
But this desire is so strong with these sociopaths, that they are willing to risk billions to achieve it.
cutting workforce is the only hype they can currently sell to business and investors. so they push that hard.
blindly turning whole consumer market against the product in meantime.
idiots managed to piss off even nerds, their main base, with the whole ram fiasco lol
very soon anything involving ai slop will be witch hunted
 
These companies pushes AI hard, but they can't explain what to do with hallucinations and data corruptions problems of current LLM models. And without solving issues corporates are vary of extensive use of AI as mistakes might be very costly.

I think most users will recoil at the thought of having an agentic OS. An AI rummaging through your private shit? No thanks.
Still, don't expect a crash, it'll find its uses. Animation, movie special effects. If you work in those fields, don't be John Henry, be the guy operating the steam-powered drill.
Crash is not like it cease to exists, but rather expectations trimmed down and investment/expansion (and a lot of market cap) along with them
 
I fucking hate a.i. but I cant fault companies for chasing it, everyone expected to get rich quick.

I know several people personally that use a.i. daily and cant even write a damn email or anything anymore without using it. Any time I see something they've "written" its a giant wall of robotic text.

Such a shame people are already getting massively lazy and rely on it way too much.
 
Copilot is bad, but they're all bad. You cannot rely on Copilot in "mission critical" situations. I have seen this myself, first hand. The idea of any company turning over operations to this junk is really funny actually because much like outsourcing, it will cost more money in the end. It's not reliable, it's not capable, it's not correct. It's useful for stuff like transcription and summarizing meetings, that's about it.

And this is matter of kind, not degree. The basic and core issues with LLMs/AI make it unsuitable for the things MS is trying to pitch it for.
 
Last edited:
RAM prices skyrocketed up to 360%. I hope OpenAI goes bankrupt

These companies pushes AI hard, but they can't explain what to do with hallucinations and data corruptions problems of current LLM models. And without solving issues corporates are vary of extensive use of AI as mistakes might be very costly.


Crash is not like it cease to exists, but rather expectations trimmed down and investment/expansion (and a lot of market cap) along with them

LLMs are Markov Chains on steroids. They really can't fix it.
 
Last edited:
I tried using Chatgpt and copilot to help me code at work and i spent more time fixing their code than it wouldve taken me to code it from scratch. im not going to say its completely useless because it does do a good job compiling search data if i want to google something, but at work it is completely useless and im tired of having execs pushing it on us. Especially when it costs so much. A collegue of mine was told to use AI to automate something but he had to explain to the execs that writing a utility to automate it would be cheaper than using AI to do the same thing.

I think executives have zero knowledge of how this shit actually works and just want to be able to say they've integrated AI with their platforms. its so stupid to watch everyone fall for this scam.
 
I tried using Chatgpt and copilot to help me code at work and i spent more time fixing their code than it wouldve taken me to code it from scratch. im not going to say its completely useless because it does do a good job compiling search data if i want to google something, but at work it is completely useless and im tired of having execs pushing it on us. Especially when it costs so much. A collegue of mine was told to use AI to automate something but he had to explain to the execs that writing a utility to automate it would be cheaper than using AI to do the same thing.

I think executives have zero knowledge of how this shit actually works and just want to be able to say they've integrated AI with their platforms. its so stupid to watch everyone fall for this scam.
Microsoft said that 30% of their code is written by AI while we watch the quality of all of Microsoft's products plummet in real time. Windows 11 in 2025 is the best advertisement against mass AI adoption you could imagine.

To date I haven't seen AI deliver anything, in a work context, that I couldn't have gotten from a Google search in 2005 before Google search turned to shit.
 
Some AI stuff is really great - speech to text, summarization, image recognition, translation etc
Some AI stuff is reasonable - better search engine, casual image generation etc
Some AI stuff just bad - complex stuff, stuff requiring memory or rare knowledge etc

The problem of AI companies is that good/reasonable stuff not worth even tens of billions and they aim for trillions, so they try to upsale stuff that obviously bad. And not eveyone buying this shit.
 
I tried using Chatgpt and copilot to help me code at work and i spent more time fixing their code than it wouldve taken me to code it from scratch. im not going to say its completely useless because it does do a good job compiling search data if i want to google something, but at work it is completely useless and im tired of having execs pushing it on us. Especially when it costs so much. A collegue of mine was told to use AI to automate something but he had to explain to the execs that writing a utility to automate it would be cheaper than using AI to do the same thing.

I think executives have zero knowledge of how this shit actually works and just want to be able to say they've integrated AI with their platforms. its so stupid to watch everyone fall for this scam.
its a glorified stack-overflow search.
useful for quick, pinpoint questions.

if you get to a point where you can't follow and double check the answers, its a cluster fuck. because it spouts glaringly flawed stuff just with the same confidence.

i honestly worry about the state of all software come next decade, some juniors will start their careers with these stuff pushed down to their throats
 
The hallucinations will probably not cease to exist entirely, btw. The best option they currently, have is to try to reduce the error margins. There will always be some percentage point for them formulating something that isn't factual or straight up fictitious nonsense.

This is what they get for leaping into this tech untested. Reckless greed overrode precaution, sensibility and reasonability. They took the opportunity and cleaned house for some ill-qualified people, but also let go of some actual good folks in the process. Short-sighted foolishness in summary.

This should've been confined in a strict and controlled miltech research domain for the time being imo.
 
Last edited:
This should've been confined in a strict and controlled miltech research domain for the time being imo.
The problem is that they try to shove AI everywhere and it's based on the same model. The paper was breakthrough, but it's a paper about language model. It really good at similar stuff - translation, transcription, domain change etc. And not so good at others where should be different models, but research ain't there yet, they simply not invented.
 
Felessan, I will politely, decline to have a conversation with you over this topic. That isn't in my interest. You can personally infer whatever beliefs you may want from that statement.
 
I'm kinda starting to wonder if its not too far fetched of an idea that the end goal of these big tech companies has become to utterly weaken people, mentally and physically, to make them hopelessly dependent on their "gracious" technology.

Tech used to be fun and genuinely served a purpose to aide and push society forward. Now its starting to look like that intent has been overturned.
That's litteraly the plot of Wall-E
 
These companies are going to find out eventually that outsourcing all of your work overseas and importing millions of visa holders to save money eventually causes your product to go to crap, including AI.
 
AI slop is not worth engaging with if you have a working understanding of the field you work in, and a functioning brain. AI slop just shows you down while having to filter out hallucinations and other completely useless noise.

Fuck MS/Copilot, OpenAI and all the other American slop factories.
 
Top Bottom