Like, I have a 'Copilot' button prominently displayed in my New Outlook on MacOS (the only platform where the app-with-that-designation is sort-of usable), and it's a dropdown menu, and it has... zero items when expanded.
I asked my 'Microsoft 365 Bing Chat AI Bot Powered By ChatGPT<tm>' about that, and it wasn't able to tell me how to make that button actually do something, ending the conversation with "yeah, that's sort-of a tease, isn't it?"...
Oh, well, and I actually also have a dedicated Copilot button on my new Lenovo laptop powered-by-Windows-11. And, guess what, it does exactly nothing! I can elect to either assign this button to 'Search', which opens a WebView2 to bing.com (ehhm, yeah, sure, thanks!) or to 'Custom', in which case it informs me that 'nothing' meets the hardware requirements to actually enable that.
So, my question to anyone in the Microsoft C-suite: have you ever tried to, like, actually use, like anything that you're selling? Because if you would have, the failings would have been obvious, right? Right??
Even in Visual Studio Enterprise, their flagship developer product, the GPT integration mostly just destroys code regardless of model output. I truly cannot fathom how any of that made it past even a cursory review. Or how that situation would last for over 6 months, but, yet, here we are.
And, again, it's fine with me: I'll just use Claude Code, but if I were a Microsoft VP-or-above, the lack of execution would sort-of, well concern me? But maybe I'm just focused on the wrong things. I mean, Cloudflare brought down, like, half the Internet twice in the past two weeks, and they're still a tech darling, so possibly incompetence is the new hotness now?
Copilot tells me it can't edit my current document, but it can create a new one. I figured okay, Microsoft doesn't want to set it loose on the original, guess it makes sense that it requires a copy. So I said yes.
Nope. Instead of creating a copy of my document and editing it, it created an entirely new document which excised basically everything in the original report and replaced it with a very short summary - I'm talking 5000 words down to 500. All my tables and figures were gone, as was the standard report template my employer uses.
What utter garbage. Office productivity is a major use case for LLMs, and here the largest vendor of productivity software on the planet is happy to fuck it up.
How did you manage this? Probably some company-wide group policy saves you. It keep starting copilot for me, drives me crazy.
I did absolutely nothing special, other than running the latest-and-greatest Windows 11 Enterprise, which is what we put on most of our laptops without any customizations other than "require 2FA and some antivirus and firewalling" via Intune.
And I just went into our Azure admin portal, looking for any AI goodies to enable, and... there just doesn't seem to be anything there? And we have an Enterprise P2 subscription, which is usually where all the good stuff is, but, yeah...
But I'd actually love to know how to achieve that, and so far Microsoft AI is awfully silent on the subject...
Satya Nadella insists that Bing365Pilot has supercharged his productivity, but determining if he's high on his own supply or lying through his teeth is an exercise for the reader.
> Copilot consumes Nadella’s life outside the office as well. He likes podcasts, but instead of listening to them, he loads transcripts into the Copilot app on his iPhone so he can chat with the voice assistant about the content of an episode in the car on his commute to Redmond. At the office, he relies on Copilot to deliver summaries of messages he receives in Outlook and Teams and toggles among at least 10 custom agents from Copilot Studio. He views them as his AI chiefs of staff, delegating meeting prep, research and other tasks to the bots. “I’m an email typist,” Nadella jokes of his job, noting that Copilot is thankfully very good at triaging his messages.
https://www.bloomberg.com/news/features/2025-05-15/microsoft...
So, his experience with Copilot agents != Average Customer's experience
I remember reading that when it first came out and all I can think is: No, he doesn't like podcasts, if you like podcasts you listen to them.
That's like saying "He loves food, but instead of eating it he feeds it to an analyzer that tells him what elements were detected in it".
I have to assume it's all BS/lies because if that's a truthful statement (about podcasts and the other things) then I really question wtf they are doing over there. None of that sounds like "the future", it sounds like hell. I cannot imagine how shitty it would be to have all my emails/messages to the CEO being filtered through an AI and getting AI slop back in return.
I guess that's worse than the Gemini button in Google Sheets that asks me to subscribe to AI services. I have multiple times been in a sheet and thought "asking an LLM how to do this thing I want to do right here in this product would actually be great if it works", remembered there was an AI-looking button in the top right, clicked it, and nope'd out of the subscription.
I just want to know if it works or not before I buy it.
Their main problem is that they never really learned how to compete on merit, just on first-to-market and all kinds of legal (and illegal) tricks.
Big Tech thinks they have a moat, when it’s really diffuse power being made available via genAI to build software good enough to replace them.
1. Stock price remains the same but revenue doubles.
2. Revenue stays the same, but stock price doubles.
Assuming all else equal, and recognizing that this is absolutely a simplification, but if these were the two choices then it seems a no brainer that you'd go with option 2. Revenue is a means of increasing stock price.
I have not looked at MS in particular, but generally that is what the remuneration of the people at the top of most public companies is most strongly linked to.
It's absurd, but that's where it is. And a company like OpenAI basically hangs on it, because they have obligation almost ten time their revenue and the only way this does not deflate quickly is if others keep feeding it cash.
Then again, I can't fathom what people would be doing with their money if the stock market weren't there. I imagine they might naturally wind up with some sort of...stock market.
Sorry but... WTF are you talking about?
It rewards self-destructive behavior in favor of short-term gains. Shareholders have *zero* commitment to the companies they buy shares from and will happily switch their entire portfolios on a whim. It's essentially people chasing the new shiny thing every single day.
Let's not forget it's a known fact that people with insider knowledge will profit over everyone else.
How is that efficient in any shape or form?
> If they undercut the US companies and are willing to accept low returns on their investments, then the respective USA competition will be driven out of business by their investors, because there will be other sectors to invest in, with higher RoI.
You're basically explaining one of the reasons stocks are a horrible idea for distributing resources.
It has nothing to do with whether or not it's central or distributed, it's merely the incentives they create. It's essentially Goodhart's law on steroids.
But would only happen if USA decided to totally financialize all sectors of its economy and make a small set of oligarchic corporations THE load-bearing element of its strategic capacity, leading us to chase market returns even if those returns totally kneecapped our ability to build anything at all of actual value.
Good thing we haven't done that!
Any empirical support for that?
It helps as it is both a gauge of the success of the strategy, and also a lever where the process can be fine tuned, eg. slowly buying stock then strategically dumping in the right time, correlated with other external shocks can have wider effect to whole industries through controlling the public opinion on specific industries.
The stockmarket enables that by making takeovers easier as you have a higher proportion of short termist shareholders who 1) fail to block value destroying acquisitions on one side and 2) jump at the chance to make a quick profit on the other.
Capitalism is defined by having the capitalist, who provides capital, and without the ability to sell their share of stock it's difficult to see what the value would be. So you kind of require stock markets.
Edit: which is why it's odd to call China communist. They have 3 stock exchanges. They're really a capitalist single-party state.
In the U.S. we have mistaken Capitalism for a religion, and so it wags the dog, so to speak. Since our founding we have made some attempts at finding a balance between our use of the tools of Capitalism and socialism (in more the Democratic Socialism style, rather than the Communism style), and we had a good run in the decades after WWII. But starting with McCarthyism, and really picking up under Regan we have prided ourselves on adopting Capitalism as a religion, and it really shows up in both the income inequality as well as the increasing role of (and corrupting influence of) money in our politics/government.
Free markets never existed, don't exist and never will. Markets are defined by laws and regulations in which they exist. It can't ever be "free".
"The term “free market capitalism” refers to an economy that puts no or minimal barriers in the way of privately owned businesses. Matters such as worker rights, environmental protection, and product safety will be addressed by businesses as the marketplace demands."
it's basically worship of owning the means of production and not being regulated in its use, e.g. if you own a company you get to dictate all sorts of unreasonable things to your employees, and any benefits gained from automation accrue to whoever can afford the up front money to own the machines.
There are better definitions on both wikipedia and Britannica:
https://en.wikipedia.org/wiki/Free_market_capitalism
https://www.britannica.com/money/free-market
Especially this hit:
https://en.wikipedia.org/wiki/Free_market_capitalism#Concept...
What we actually see is a system of chartered extraction. Corporate executives are like Norman lords, granted their 'title' (CEO of instead of Earl of) by shareholders (rather than a king) in return for which both are/were expected to extract maximum value by any means necessary. Extractive tactics often at the expense of long-term product strength are behaviors shareholders expect if the CEO is to keep their bestowed 'title'.
Don't forget the progenitor joint stock company The East India Company, Capitalism in it's purest form without government restriction. Profit-maximizing, absentee extraction, with company executives serving as quasi-feudal lords over assets and people. Modern corporate capitalism is hard to distinguish, in its structure,history, behavior, and incentives, from the Norman extraction system, it's just dressed in a more politically palatable wrapper and forced to mellow out from it's desired East India Company style final form.
The only good thing that came out of Satya era has been the Windows Terminal and WSL.
The sad reality seems to be that Microsoft do not care about the majority of their products anymore. Only Azure, Microsoft 365 CoPilot, CoPilot and maybe CoPilot.
Microsoft 365, which I believe includes Office, makes up $95B of that amount, which is split between Commercial (92%) and Consumer (8%)
From there you can see why they're focused on Enterprise.
Source: https://www.bamsec.com/filing/95017025100235?cik=789019 (page 39)
It doesn't make it any better that Microsoft does this, but as a piece of practical advice, it seems like it can be done. There does still exist a core of Windows under all that garbage that is fast.
Honestly, I had to do a lot of workarounds to get comfy. There's annoying stuff I cannot uninstall.
If you haven’t seen how bad it is, here’s one example: Copilot Terminal. In theory, it should help you with terminal commands. Sounds great. In practice, it installs a chat panel on the right side of your terminal that has zero integration with the terminal itself. It can’t read what’s written, it can’t send commands, it has no context, and the model response time is awful. What’s the point of a “terminal assistant” that can’t actually assist the terminal?
This lack of real integration is basically the core design of most Copilot products. If you’ve been lucky enough to avoid them, good for you. If your company forces you to use them because they’re bundled with a Microsoft license, I genuinely feel your pain.
Their inability to produce anything useful with Copilot is the largest example of this, but there are others. They are getting lapped by a ~300 person software company in the race to consumer-ize an x86 PC a into turnkey gaming platform, even with $100 billion in game studios and owning the API that every major game is developed against. Their footprint in education is gone, completely replaced by Google who not only produced an operating system that could be effectively run and managed on commodity hardware, but also developed the centralized functions for school administrations to use to manage classrooms at scale.
The consumer situation for Microsoft right now might be even worse than it was when Nadella took over.
You say this like it was a mystery to start with. When you own 90+% of the user base, you can create trends with any changes implemented
Find a software market currently addressed by high price products; create a reasonably good product for that market; sell it for significantly less than the incumbent. Sell much higher volume of said product than the incumbent, thereby make much more profit. Repeat/rinse.
The Windows lock-in, embrace extend etc came after this. You can't lock in customers if they didn't already willingly buy your product.
From 1981
>Microsoft, which needed an operating system for the IBM Personal Computer,[9][10] hired Tim Paterson in May 1981 and bought 86-DOS 1.10 for US$25,000 that July
It has been an incredibly lucrative strategy. We all herald some CEO's prowess in growing revenue when they've been doing the same playbook for decades now, and have been running on the inertia of Windows dominance on the desktop. Every new entrant is pushed out through countless incredibly lazy IT departments that just adopt whatever Microsoft shits out.
It's actually surprising that the one and only area where this really failed was as they tried to lever tying to the mobile market. A couple of missteps along the way are the only reason every office drone isn't rocking their Lumia ExchangeLive! CoDevice.
Google ships features people actually use; Microsoft ships demos people tweet about.
In AI, “ship it now, fix it later” doesn’t work when everyone else is shipping things that already feel finished.
Can't speak to the part about Microsoft, but it's obvious Google is creating AI products the employees want to use and do use.
From my experience, Microsoft’s GPT-5 integrations in Word, PowerPoint, and their ChatGPT clone struggle with basic tasks. Copy/pasting from ChatGPT still works better.
To be fair, building solid AI features is hard when model capabilities change so quickly. Reasoning and tool use only became reliable in the latest models, and when these Office features were planned, GPT-5 didn’t exist.
Side tangent: Copy/pasting from the Windows Copilot app is absolute dogshit. It makes no sense that the simple action of "copy this text" is this broken.
What a bizarre way to organize the chart. Clearly Anthropic is leading -- their early bet on "programming" as the main use case is paying off.
The recent report from Openrouter [1] confirms as much: coding is the number one use case, with role playing / fantasy writing in second place.
I have a feeling the remaining use cases will never dominate, instead they will slowly mature into acceptable practices across the various industries. That will probably take longer than the investment bubble can hold though.
Satya’s days are numbered. The OpenAI investment will be a very costly and painful lesson.
When the internet came out, did many legacy companies lead the way with online experiences, figuring out what the real killer apps now that everyone was connected were? I don’t know for sure, but I doubt it, I think it gave rise to some of the present crop of big tech, and others reinvented themselves after the use cases were discovered.
All that to say, I expect the same here. In 10 years there will be AI uses we take for granted, built by companies we haven’t heard of yet (plus the coding apps) and nobody will talk about stupid “rephrase with AI” and other mindless crap that legacy companies tried to push.
Their investment in OpenAI, giving them what was, at least ~1-2 year ago if not now, the best possible LLM to integrate in the office suite yet they are unable to deliver value with it.
Their ownership of Xbox and Windows should have allowed them to get a much better foothold in gaming yet their marketplace is still, to this very day, a broken experience with multiple account types. It's been 10 years.
The counter point is Azure obviously which still has great growth numbers, but that's a different org.
From the outside, it just seems like they should be doing better than they are. They have much better business integration than Google and Amazon. The fit is obvious and people are borderline hooked on excel. Why aren't they dominating completely?
Nadella has done a lot of listening through is CEO reign but it looks like MS is back in a "don't listen to customers, tell them what they'll get" phase.
On iOS, you type a message and send, and... nothing.
Although knowing Canonical they might add something to Ubuntu sooner or later.
Normally you get a frontier exploration phase where fringe people experiment with the new technology and try to figure out what it's good for. It feels like we just skipped that step entirely.
Every procurement team is going to point to copilot, saying it’s included with the other Microsoft services a company is already paying for, so duplicate AI products won’t be approved for purchase.
Microsoft is laying claim to the desktop real estate, so in a few more generations of the technology, they’ll have the customers and competitors will already be starved out.
Right direction, wrong execution.
However; their azure offerings are somewhat decent. AI Search is getting quite decent, access to Open AI models served through foundry is quite the differentiator (although the rate limits are an issue), their reference architectures are very helpful and their ancillary services, like document intelligence integrate well.
I'm a bit worried that their marketing dept has gotten their crappifying and obfuscating eyes set on those services though, given the recent rebranding of Foundry and what not, but the underlying services are decent.
Updated as it was almost close to being a generic comment about AI overall.
And noone should actually be shocked about his ineffectiveness. Covid was a great example of how clueless his leadership has been. Skype used to be a verb people used in common parlance, and yet they dropped the ball and let Zoom take over both consumer and enterprise segments while focusing on "restructuring" Skype into Teams for no reason whatsoever.
Prior to Covid, he was ready to let Windows run its course and axe that too. The sudden demand for sub-$500 laptops during the pandemic showed him that people still liked Windows and wanted a good OS from Microsoft. But instead of capitalizing on it to give customers what they wanted, he just gave us an ad-filled spyware with AI slop.
I have zero hope in any product with a Copilot in its name (including GitHub). At this point, unless there's a change in leadership, it's only a matter of time before XBox faces the axe.
Edit: Or better still, convince all of their customers to throw away perfectly good hardware and upgrade to one with a single extra chip, creating a hazardous waste epidemic for landfills as a nice side effect. It's especially important to do this in the middle of a RAM and HDD shortage.
Really, I'll just never be half the great business strategist that these guys are. <sigh>
That monkey face simply won’t go away.
ChatGPT respects privacy and allows for maintaining history while also opting out of using ones data for model training
And I don't trust Sam Altman and AI.com at all since their whole thing was built on lies. They could start regaining the trust by changing their company name.
Yup, yeah, sure. The company that attempts to open your password-protected zip files. Let us not give it a free pass either.
There is no good incumbent.
This reads more like a hit piece than good faith article
(But yeah the MS AI products especially on consumer level are pretty terrible)
It is going to be very hard for OpenAI or Microsoft to compete with this now that Google has gotten their act together.
It's not just AI, it's a market fit and quality problem.
They don't need to solve it, however.
Their strategy has been quite clear: make it barely usable so that is passes muster to auditors, integrate it with systems that corporations need, and sell them on the integrations.
Teams and Azure suck?
So what?
Big companies will pay for that, because it's integrated with their ldap, has an audit trail, gives them the ISO-whatever stamp, and lets them worry about something else.
That the users are miserable is almost never the question for the ones signing the checks.
In a world where box-checking is paramount, this approach is a winning strategy.
Helpful chart to draw conclusions
But this also means they end up with average products. They don't have the talent to do something exceptional.
This has worked well for them when they can just come in and copy something (say AWS in Azure) and not pay the innovation cost, but AI seems different for some reason, perhaps in the same way search was. You need the top 20% in order to really be successful.
Plenty of SV is building below average products with exceptional talent.
This is where they are failing.
> Plenty of SV is building below average products with exceptional talent.
Yes, you can hire exceptional talent and give them poor directions, resulting in poor products.
But to hire mediocre talent and still produce competitive products you must have an unfair advantage of some sort. The Windows and Office monopolies gave Microsoft that unfair advantage. But it is becoming clear that this unfair advantage does not extend to AI.