It’s partly that, but it’s also partly that the quality SUCKS. I’m frustrated with AI blogspam because it doesn’t in any way help me figure out whatever I’m researching. It’s such low quality. What I want and need is higher quality primary sources — in depth research, investigation, presented in an engaging way. Or with movies and shows, I want something genuine. With a genuine story that feels real, characters that feel real and motivated.

AI is fake, it feels fake, and it’s obvious. It’s mind blowing to me that executives think people want fake crap. Sure, people are susceptible to it, and get engaged by it, but it’s not exactly what people want or aspire to.

I want something real, something that makes me feel. AI generated content is by definition fake and not genuine. A human is by definition not putting as much thought and effort into their work when they use AI.

Now someone could put a lot of thought and effort into a project and also use gen AI, but that’s not what’s getting spammed across the internet. AI is low-effort, so of course the pure volume of low effort garbage is going to surpass the volume of high effort quality content.

So it’s basically not possible to like what AI is putting out, generally speaking.

As a productivity enhancer in a small role, sure it’s useful, but that’s not what we’re complaining about.

> sneaking its way into where it's not wanted

This. After a generation of social media sneaking its surveillance, manipulation, and noisy ads into our home, work and mobile lives, it is very obvious that having something "smart" shoved into tools where it wasn't asked for isn't some noble attempt at improving lives.

Users are tired of being continually and transparently abused.

All Microsoft would have to do to shock the world and get months of good press is announce they were never going to opt anybody into anything by default any more. At this point that would be considered astonishing.

And suddenly, internal incentives would be to create useful, conflict-free capabilities users actually choose for themselves.

Sex is great, but if you constabtly try to force it on me, sneak it into deals we make, or while i sleep etc. it will leave me quite hostile towards the topic, and that's where we are at. Consent is important in all things, no less here.

This is where a lot of people are. In my case, every time I open a PDF in Google Drive, it forces an AI summary of the doc with no way for me to switch it off. I try to close it mid-generation, but I suspect not fast enough to keep it from getting counted in the usage stats, which is probably what some product manager is trying to maximize and demonstrate ("X number of PDF summarizations this quarter").

It's also not very good at any of those things, if you ask it to generate something far enough outside of the mainstream, or something particular, or something consistent, or- But, yeah, the insistence that we deprecate every other even remotely-connected resource (including other people) in order to supplicate ourselves to corporate desires is aggravating. You got a lot of the same pushback with VR. VR is really, really cool. Having your reality mediated by large corporations with a history of user-hostile behavior is not. Them not taking no for an answer feels violating.

Reminds me of the classic in the genre:

  how about a couple of weeks of gratitude for magic intelligence in the sky, and then you can have more toys soon?

  Sam Altman, Sep 12, 2024

This is the fundamental misunderstanding that the AI techies have. Normal people don’t want a “conversation“ with a computer. Conversations happen between people. Computers, at best, receive orders and carry them out. They are tools, not companions, and they should do nothing unless explicitly told to do it.

I would be impressed if AI was actually 'super smart' but what we have now is not.

> It's not that people are unimpressed with AI

Oh no, I am definitely unimpressed. That AI you can have a sorta-kinda fluent conversation with is often a complete moron and a habitual liar, and the images it generates are awful - did he not see how horrible that Coke ad looked?

It'll probably end up useful in a bunch of applications soon-ish and I'll probably want to use it eventually, but in the meantime their AI is flooding the internet with absolute garbage, and they are literally shoving AI in my face at every opportunity they get.

It is painfully clear that people just aren't that interested, and they are getting increasingly desperate about finding ways to recoup their massive investments. But people aren't going to magically become enthusiastic about eating rotten garbage if you just keep stuffing it in their mouth!

If anything, their current approach is only going to make people hate AI even more. But they are in too deep, and admitting defeat and scaling it down until they have an actually good product that people genuinely want means seeing their stock price crater because they will have "lost" the "AI race". Their only option to avoid an immediate collapse is to keep lying through their teeth and keep trying to pretend that it is absolutely amazing and that you just must use it.

Or maybe the CEOs are completely delusional and genuinely believe what they are selling - I'm not sure which one is worse.

What a gaslighting king this CEO is, the concern isn't how functional the AI is, the concern is AI just downloading any and all PERSONAL AND PRIVATE files on a whim, with no guard rails. What if I have photographs of my kids I've never uploaded anywhere EVER because I don't want them anywhere outside of MY DEVICE, does Microsoft just magically get to suck in those files and own them? Wild.

Its this shenanigans that forced me to nuke my Windows install and go Arch. I noticed that Windows Defender will upload "suspicious" files and there's no audit trail of what's being uploaded. So I have no way of knowing what personal documents or even proprietary software has gone up to their cloud.

The fact that people are unimpressed that we can have a fluid gameplay experience of the latest entry in the Diablo franchise in the palm of our hands was mindblowing to someone as well. "Don't you people have phones?"

Instead of "pushing back" I wish he'd actually listen to and address his critics' points.

He is deliberately and disingenuously missing the point. It's not that the features aren't good (maybe they are, maybe they aren't). It's about how coercive Microsoft and Windows are with its users, and this exec is failing to address that one.

Just once, I'd like to hear a question get through to these assholes asking them why they are forcing so many unwanted things onto their users. From Microsoft accounts to forced windows updates to Recall... Gone are the days when users had any control over what their computers are running.

But these kinds of questions never seem to get through to them.

Exactly. It's not that everyone is saying "AI is completely worthless, get rid of it." It has it's use cases, I certainly benefit from LLMs in my job every day.

That doesn't mean I want it plastered everywhere, in every app or website. That doesn't mean I want to interact with or use my computer via AI, and I especially don't want to talk to my computer to do things. Mouse & keyboard is faster.

But for now at least you can just choose not to use it. The problem is, Microsoft is putting 100% of their efforts into this while long-standing Windows bugs and regressions still exist. They're aware they exist too, and are deliberately choosing not to improve their product.

I have never in my entire life wanted to "generate an image or video". I like taking photographs and recording videos because they represent the reality of my life. Who would ever want to "generate" fake images as a matter of normal daily activities?

"we can have a fluent conversation with a super smart AIwe can have a fluent conversation with a super smart AI"

But we can't. I can have something styled as a conversation with a token predictor that emits text that, if interpreted as a conversation, will gaslight you constantly, while at best sometimes being accidentally correct (but still requiring double-checking with an actual source).

Yes, I am uninterested in having the gaslighting machine installed into every single UI I see in my life.

Exactly. Even if we grant what he says (I don't fully agree)—that doesn't warrant putting that kind of "conversational engine" in Windows as a first-class citizen.

I don't want to have a conversation with my computer about my Word docs. I just want to write my Word docs.

I don't want to have a conversation with my computer about the quarterly report. I certainly don't want it making up values for the quarterly report. I just want to write the quarterly report.

Having a conversation with a computer is cool. It's a fun party trick. If there were a way to reliably get it to know about all of my things, without the concern that it would then take all that data and feed it to its mothership, I might want to be able to converse with it about those things, under certain circumstances.

But, yes: if I want AI I'll actively seek it out and use it. Stop acting like me being upset that it's getting shoved in everywhere is the same as me saying "this is a meaningless achievement."

When have i ever wanted my os to “generate any image i want!”

This is like a chef being confused why people dont like the shoes he made them. Why did he make hungry people shoes? Certainly not to eat?

I use AI everyday and it’s now integral to my workflow. However, even I still hate the hype train and having it constantly stuffed down my throat. Nevermind AI slop.

Windows 11 is already adware. No wonder people are complaining about more ads.

I am unimpressed with it. If I wanted to steal code off stack overflow I can do that myself. Another layer of indirection has negative value.

I can generate images that are difficult to use commercially. I can analyze something with AI but I can't confidently use that output in any setting that matters.

For people who are attempting to engage in profitable work then AI is miserably unimpressive. I don't know what planet this guy is living on. Time is money. Flowery emails and off axis summaries can only create a waste of that time.

Well, and companies are papering over usability problems with AI. LLMs are not a substitute for good human-centric design.

It's almost as if all the focus has been on eliminating the human... for products designed for humans.

"If I want AI I'll actively seek it out and use it - otherwise, jog on."

If I want MS Windows I'll actively seek it out and use it - otherwise, jog on

If this is not a statement you can make, then Redmond gets to decide what you use, not you

That’s a great quote —- when you hire a “creative” to do a job (ie photographer, designer, etc) in area where you aren’t an expert, the conversation of what you want usually starts with similar existing work that you like.

A good creative will take that as a starting point, apply their skills and vision to it, and give you something that solves your specific problem in a unique way, often far better than whatever you had imagined.

In my experience, if you do a similar expertise with AI, it just gives you a facsimile of the inspiration you initially provided and not much more.

I'd love to read more wherever this came from if you could link the source

Don't forget the audience member who literally asked if it was a joke - and got cheers and applause from the rest of the audience. It was probably one of the biggest PR disasters in gaming history - and it does seem like the AI CEOs have been taking quite a bit of inspiration from it.

My thought exactly. From hindsight, Diablo immortal is not a bad game, but that moment was really…not great. I guess the guy knew that phone games were getting momentum but unfortunately that specific group of users in Blizzcon didn’t want a phone game.

Windows is not a general purpose operating system. It's a platform for monetizing businesses via licensing and cloud services, and a platform for monetizing private users by way of advertising and data mining.

If they were ever to produce a Windows PowerUser edition, with absolutely no bloat, it would have to be priced like a CAD suite.

"AI" is not a machine for improving the lives of people, it's a tool for consolidating power and wealth at the top. Declining welfare and access to basic human needs shows that decision makers are getting better at min/maxing the economy.

> AI failed at Microsoft...

Let's be fair. Microsoft has not succeeded at mainstream consumer AI products... yet.

To say AI failed at Micrsoft when CoPilot (the real one, for developers) was, last I heard, the most subscribed generative AI tool for software developers is not a failure. It's wild that most developers in a corporate environment pay Microsoft to use Chat GPT, Gemini, and Claude. Their other gen-AI components in Office, like Powerpoint and Word, are pretty darn good. But again, unless you're a corporate user in a work setting, you probably don't care.

This push to lease you your own computer is what hasn't worked very well so far. I dearly hope it pushes more people to Linux (though more likely they'll flee to Mac, which is a more palatable version of the clumsy crap MS is trying to do).

So consumer AI... perhaps that has failed. But the money isn't in you and me paying for a Windows license. The money's in big corporations paying for ten thousand seats at a time for their suites.

it's pretty good at highly specific questions about software support, from my experience. I'll say what program I'm using, what I'm trying to do, and what errors I keep hitting.

"Click this, then that, then this other thing and it should work"

"that other thing isn't an option"

"Oh you're 3 versions behind. Instead, it's in location X."

This is one thing that I miss from Balmer days.

I mean, if they want to put a grifter and a con artist in charge of their division and torpedo themselves, I say let them go for it!

Something, something, never interrupt your opponent whilst they’re making a mistake.

Considering Diablo Immortal generated a ton of money, I hope you are wrong.

I believe users are stupid enough to stick to Microsoft "agentic OS" anyway.

For me it’s the opposite. AI has enabled me to write songs and create pictures. I can’t draw but AI helped me to implement ideas I had.

Reminds me of the Xbox One reveal disaster.

Because they're right. They'll all collude to make sure that you can't install a non-"smart" OS, just like you can't buy an non-"smart" tv, and the public will be well-trained to mumble something about the "market" and call you names ("luddite," "techie," "boomer," whatever...) when you complain about it.

Once a generation has been raised who never saw a computer that couldn't refuse to let you type what you wanted into it, young people will stop believing that you could ever type what you wanted. Old people will forget that you could ever type what you wanted. It worked with literally everything else.

If it weren’t so tiring having to wade through all the ai slop they add to products I’m forced to use, the fact that leadership in the field are getting petty and childish about how not everyone thinks their toy is shiny is a real “telling on themselves” moment.

If it actually, truly, world-changingly good as they are _begging us_ to believe they are, they wouldn’t need to care that people disliked it or chose not to use it.

But because they’re practically going red in the face screeching about it, it really comes off as “cope”, to use the hip new word.

I'll take Charms, mouse gestures, and the Start Screen over Copilot any day.

Heck, I Do have day-to-day use for it, and I still don't want it to completely take over how I use and interact with my OS.

Nor do I ever want to have a voice conversation with my computer to where it responds in an uncanny valley voice. If I do want to use voice, it's to give a command. No response needed. "Hey computer, call John" that's it. Do the thing, don't talk back. A glorified voice assistant is all the further it needs to go.

It's not even a question of trust or loyalty. Microsoft has explicitly told us they're doing bad things with the data.

I don't think Microsoft as a corporation is any longer even aware of customer trust as a concept. All we are is a KPI and a credit card to be exploited for anything they feel they can take.

We aren't even people to Microsoft anymore. Just a revenue source to be maximally exploited right now with no concern for future revenue.

> Microsoft also doesn’t have user trust in the way Apple does, so everyone just assumes MSFT is going bad things with the data.

I think this is a big part of it. If Apple ever achieves their vision for personal context Siri & AI in their OS, I bet people will praise it and actually use it. Because Apple has built trust with their customers, and has strong marketing around privacy.

Microsoft burned that bridge a long time ago. They feel sleazy. Maybe if they haven't violated their users trust over and over again, people would be more receptive.

Microsoft has been pushing AI features into both VS Code and GitHub for quite some time

This made me think of Henry Ford's quote about people wanting a faster horse. Granted, he was an extreme outlier, which is why he's worth quoting

I'd bet good money that he thinks all his exes are "crazy"

If Elon musk bent over and sucked shit from his own ass and Tesla stock went up, every CEO would start stretching.

Unfortunately Apple seems to be gong the same way. On Mac and iPhone I am seeing more things that don’t work well.

Meanwhile we get a react native web app for a start menu, and you can't even move the taskbar's location on the screen. A feature that has existed in Windows since 95.

Something is seriously dysfunctional in Microsoft.

That's gotta be it.

At Ignite yesterday they announced that Security Copilot will now be included for free with E5 licensing.

The tool that until yesterday way $50k for a single tenant deployment. Aka, no one bought it, but they need to juice the KPIs so might as well make it free so it looks like someone is actually using it.

On the topic of grammar errors. I have seriously considered leaving several grammar and spelling errors in my posts in order to provide a signal that this post was written by a human. And then I get several sorts of depressed at what this implies for our future.

He grew up in Britain so he's most likely a native speaker.

It's a Twitter post, not an English essay that's about to be graded.

I'm a Brit and the grammar seems ok to me. I would have put a comma after the jeez. Not sure about the content though.

Seems like he should've used Copilot, is own freaking product, to make the post for him.

Other than missing the word "are", are there other grammar issues with the post I'm missing?