CyberCPU Tech

The BETTER Windows 11 Pro! youtu.be/9Zd58dgys28

@Lycancub940 @fandompulse Okay. But you didn't even touch my main point. You'd have to hire a large percentage of the Earth's population to moderate a platform of 3 billion users without automation.

@cybercpu @fandompulse And yet the A.I is continuesly making mistakes while said people are going oh its perfect its never wrong. Blinded by A.I yet us as humans are becoming brain dead.

After demonetizing a load of creators, YouTube sees a double-digit uptick in ad revenue in 2026, generating $9.88B of ad sales in Q1 (up 10.7% year over year). The platform generated more than $60B in revenue in 2025, according to Alphabet, making it the industry’s biggest winner in terms of entertainment revenue. Is this why they're nuking so many creators so they don't have to pay out?

@DrGravityX @nailath_vt 😅🤣😂

@cybercpu @nailath_vt (the exact policy page I linked) and for inauthentic/mass-produced content (the definition you quoted). Those are measurable, policy-level definitions. ur absolute claim that no such category exists is false. Provide evidence for your original two claims, or concede the error.

My newest conspiracy theory: Either Youtube's "inauthentic Content" Bot is faulty and to aggressive... OR Youtube is in financial trouble: - they reduce the requirements of partners from 1000 to 500 Subs. - they increase the amounts of Ad's. - they make content creators pay for their own Video promotion. - they add Youtube Premium to reduce Ads. - they use an aggressive Bot to cancel Partnerships to avoid paying Creators, because a bunch will just give up or have no way to appeal the decision to get earnings back. Tell me im wrong.

@DrGravityX @nailath_vt AI slop is destroying YouTube. There's only one reason to defend it.

@cybercpu @nailath_vt “You claimed these two statements ‘don’t even fit into [the scientific method] category.’ That’s a universal negative. you’re saying there is no objective framework for either. But YouTube itself has an official operational definition for both ‘enhancing storytelling’

@DrGravityX @nailath_vt The only problem is that my opinion is shared by the majority. People hate AI slop.

@cybercpu @nailath_vt "the only times i see that not being the case". exactly the point - it's what "you" saw and that's a biased sample. that's an overgeneralization error. your using what you "saw" and trying to generalize it. this is an error. but as you said it's your opinion so yea..

@dylanweenter @WindowsCentral And blocked. 🙄🙄 Simple bugs can be fixed in 10 minutes. Like not being able to click on a desktop icon. 😅😅

And if they would have fixed whatever control panel bug you're talking about you'd be complaining about the task bar right now. But for the last year I can't open folders from the KDE desktop without clicking like 300 times on them. But it's okay, Linux can be buggy as hell and it's cool. 🙄

🪟Windows 11 May update on the 12th looks like a GOOD one! The May 2026 Windows 11 update is one of those releases that looks small on the surface but ends up changing how you use the OS every day. We walk through 11 new changes that you'll want to look out for when it starts rolling out. From our Windows expert Mauro: "I walked away from the May 2026 Windows 11 update feeling like Microsoft finally struck the right balance between useful upgrades and everyday polish. Xbox mode is the feature that grabbed me first because it genuinely makes my PC feel like a console when I want it to. File Explorer feels faster, folder views are finally consistent, and the expanded archive support saves me from installing extra tools. Even the small touches, like smarter Taskbar AI and tighter driver rules, make the OS feel more intentional. It is not a flashy update, but as I tested everything, I kept noticing little improvements that add up." (1/2)

Well, you see that's another situation where you're not being fair. So you claim Microsoft has more resources, therefore windows should be more stable? Between 3-5k developers work on Windows while on Linux there are between 11-25k developers contributing to the kernel or different distributions. So it's like five times as many people working on Linux than Windows yet Windows is supposed to be more stable? Listen, I think there's positives and negatives to both, I just find it funny how people are hyper critical about issues in windows 11 that typically get fixed fairly quickly but completely silent on issues in Linux. Like for instance, for the last year my KDE desktop won't respond to clicks to icons on the desktop 60% of the time. Most of the time I have to open Dolphin and then go to the desktop folder in order to launch something. It's ridiculously annoying. Then for the longest time, every time I updated my kernel it corrupted my video driver and I didn't typically realize it until I restarted the system and it wouldn't boot into X. So I would have to reinstall the Nvidia driver from the console. It's stupid crap like that people just brush off their shoulder like it's no big deal but if there's a table out of alignment somewhere in Windows it's the end of the world. 🙄🙄

@cybercpu @WindowsCentral And a person who had no money created Linux without all that money, and most importantly, the Linux kernel is open source, while Windows isn't because they want a monopoly.

Are you sure that they don't want to and don't listen to their users? Because regardless of their faults Microsoft is doing a massive rework of Windows 11 directly because of complaints from users. They're optimizing the UI and making a bunch of massive changes. So they're currently, right now, In the process of doing what you're criticizing them for not doing.

@cybercpu @WindowsCentral I never said there weren't any. The only absurdity is that a company with billions of dollars doesn't want to and doesn't listen to its users, doesn't optimize its product for better performance, which is used by billions of people all over the planet.

@dylanweenter @WindowsCentral What gets me is how hyper critical Linux users are about Windows while being very uncritical about issues in Linux. Linux can be littered with bugs and it's cool but one tiny detail in Windows isn't working right and it's the end of the world.

@WindowsCentral They're throwing dust in people's eyes like they're doing something better, but in reality, these are the most insignificant winterizations that should be according to Default. If you're on Linux, Windows is still complete nonsense.

@FeltSteam @flowersslop @swans11000 @leecronin This is what they're talking about when they say that AI psychosis is it departure from reality. You're trying really hard to claim AI is alive without claiming AI is alive. It's nonsense. I can't wait for this bubble to pop.

@cybercpu @flowersslop @swans11000 @leecronin Actually addition is a perfect example. The models have come up with their own logic that we didn't program in to handle addition, its a fairly complicated multi-step thinking process that happens within a forward pass but I would recommend looking into it #dives-addition" target="_blank" rel="nofollow noopener">transformer-circuits.pub/2025/attributi…

@FeltSteam @flowersslop @swans11000 @leecronin Ok, whatever you say. You can either make a choice. You can have an internet full of content or an internet full of bots. You can't have both. Choose wisely because the bots use the content to train so in the content disappears so will the bots.

@cybercpu @flowersslop @swans11000 @leecronin Ok I went through the first 7 minutes of the video and there are a number of misunderstandings, I give a few examples here pastebin.com/3A0BPy4f

Okay, whatever you say. If the AI bubble does not pop and AI continues the direction it's going there will be no content creators. There will be no new content and with no new content there will be nothing to train the AI then the AI will fail as well and it will have taken the entire internet with it. That's our future. Hopefully the AI bubble pops before that happens because if it doesn't we will be living in the '80s again.

@cybercpu @leecronin You’ve moved from “it’s just copying” to “it has market and licensing issues.” Those are not the same claim. Verbatim reproduction matters. Creator control matters. But that doesn’t make normal outputs copies, and the legal question isn’t settled. 🍪

They are hyper specific AI models that are following rules to come to specific conclusions. Kind of like when you type 1+1= into a calculator it responds with 2. The calculator doesn't actually understand numbers. It's been programmed with logic, by a human being, to come to specific conclusions based on rules. It's not thinking.

@cybercpu @flowersslop @swans11000 @leecronin Sure, I'll watch through it now. Though I would be curious on your thoughts of them solving Erdos problems.

@FeltSteam @flowersslop @swans11000 @leecronin Yeah, I listened to you. You're just wrong. LLMs do not have the ability to use logic and reason. They can mimic it, but they themselves cannot think for themselves. THEY ARE NOT ALIVE.

@cybercpu @flowersslop @swans11000 @leecronin

@FeltSteam @flowersslop @swans11000 @leecronin And here we go, I just don't understand how LLMs work. 🙄🙄 Here, I actually did a video breaking down how LLMs work. Why don't you tell me where I got it wrong. youtu.be/NWQ1CFIj8zo

@cybercpu @flowersslop @swans11000 @leecronin Believing all LLMs are is stochastic parrots is just being ignorant to how they work, plus if models had no capacity to do any form of logic or reasoning how do you suppose they can solve unsolved Erdos problems with almost no human input github.com/teorth/erdospr…

@FeltSteam @flowersslop @swans11000 @leecronin No, they don't. LLMS don't even understand the prompt you type in. They're just really good at mimicking logic from their training data.

@cybercpu @flowersslop @swans11000 @leecronin Except LLMs do understand, do logic and reasoning. Look at Anthropic's interpretability papers, but there are other cool ones like arxiv.org/abs/2507.08017

Dude, piss off. I am so frustrated today with half of my notifications coming from people defending AI misrepresenting me and then the other half of my notifications coming from people like you claiming I've defending AI. I can't win either way. I've already explained myself. Read moron.

@cybercpu @nailath_vt You dont want to explain yourself, because "i already have". If I was wrong in my statements, you'd be able to easily prove me wrong, which you cant, because im right. FYI even if humans have an error rate, you can at least reason with a human, you cant with AI. Clanker lover

I've explained myself in detail. I don't feel like going over it again. If you want you can scroll up and read my argument above. Just keep in mind though, human reviewers are going to have a false detection rate as well. In fact typically the way social media platforms do this is they outsource moderation to third world countries who have to keep quotas. So the moderation gets just as bad if not worse than AI.

@cybercpu @nailath_vt Then why dont you explain what your argument is, because from what ive read, youre defending YouTubes ai and saying human reviewers arent feasible

@flowersslop @FeltSteam @swans11000 @leecronin @grok This is just proving my theory. The only people defending AI are those who are financially tied to it or lack talent to be able to perform without it.

Well, before AI psychosis I would have thought that answer was obvious but apparently it's not now. Because human beings have the capability of logic and reason and as sentient creatures we have the ability to learn. LLMs don't. So if they're memorizing entire sections of copyrighted works then that data has to be stored somewhere.

@cybercpu @flowersslop @swans11000 @leecronin If you broadly say "An LLM memorising its training data is in fact proving that it can act functionally like a database" why doesn't "A person memorising a book is in fact proving that they can *act functionally* like a database" work?