Logbook: Computer hardware
My personal stance on AI.
AI can be a great and terrible thing. But, I feel like AI in its current form is crap, companies trying to shove it into everything possible like AI lawn mowers. Why stick a computer in a lawn Mower that tries to use GPS that ends up killing your neighbors roses when you can use markers on your lawn. AI coffee pots? no give me a power button damn it. Web browsing is been enshittified to make the AI browsing more “effective”. In the past you could search something on google without 10 pages of garbage because the search results were vetted, Checked and than indexed by computer.
The problem with AI is it is centralized, We have to ask one machine. We have to ask one machine to talk to another machine to talk to the software that talks to another machine that turns on a light bulb. It is this fragmented centralization we pay the devils due to. By saying hey _____ turn on the light, The machine took your input, checked your associations figured out which company owns the light bulb , gave up your data , gave up usage and likely sniffed your network just to make a 5000 mile trip halfway across the world to turn on your light bulb less than 10 feet from you. By the time you weigh out your privacy cost your cool colorful light bulb sniffed your network or your bluetooth and found you have a bluetooth vibrator in your house. The app that controls your light bulb is now giving your personal massager ads now.
Now that i firmly have shit on centralized AI, I need to make the opposite argument for AI. Having a deeply centralized machine to an intuitive person can be amazing. Research that was done with hours of pouring over google, bing, Yahoo because they all index differently, with a side dish of wikipedia’s articles with the comments on wiki took hours, Now with AI you can ask the question have AI either excerpt it or in my case show me point of view that conflict each other to get a more whole perspective on thing is great. but, There is one caveat, Vet your research, do not assume AI is always right . Just like a librarian your helpful AI will bring you boundless information on your subject of research, but if your AI librarian gets confused just like the human can give you output that makes you go what the fuck? But if you properly vet your research (meaning: check its work) with the AI , It can find information that before would take hours between 3 search engines 1 online encyclopedia and 3 cups of coffee, you spent your day researching on the failures of the “streaming industry” and you’ve barely started your work. Now with AI you can work AI as a vetted peer researcher, you can tell it when it is wrong. The websearch that was the past took know how of using the web search keys like “” – + that 99% of people do not use.
Where my final thoughts here is… Do we need a centralized AI? Yes and NO. Why because centralized information for an AI makes for great research. But does my home device need to connect to that to turn on a light? Fuck no , that device should use a cut down version of the AI locally that only knows how to turn on lights , Adjust your heat and the other simple joys around the house, If you have a AI coffeepot or teapot Call me when you can do “Tea Earl Grey hot” or “coffee whole milk, semi sweet”. Only when decentralized AI does not understand the query should it ever “phone home” .
Sometimes, with AI the same goes for image generation, It is useful, but right now its just massive shitposting. If I as a photoshop user want to save several hours making an image I will annotate to the AI the image I want but I will not take any claim to it . I wont hide the tagging Gemini puts on the image because it is a time saver to me. And a lot of the times I let the image generation do what it wants because sometimes its funny as hell to watch the smaller hallucinations of a peer check on an article play out in the image. In my life if an AI saves me an hour creating an image I will let it come up with something. In my real life with my canon camera I will never let AI touch an image I take, I prefer nature and the perfect chaos that real life is to capture the best image. I prefer a natural smile to an AI “fixed” image. they look plastic.
So while you may read into my visions to AI as hate , its more like critique for a better world where information is not sold but given to make us better as people. Otherwise whole saleing information behind locked doors just makes us look as bad as the 1100s.
This post is long and if you have got this far without AI summarizing it for you, Enjoy your next sip of coffee and give pat yourself on the back. Im proud of you.
Meta destroys the AI powered worlds to make the LLM world.
It would seem like Meta is the first to throw in the towel with their announcement of the Meta horizon worlds is going to be retired as of June, This was a platform powered by slop. The Idea of building a universe with AI was a bad idea in general. WIth that you would end up with places such as 6 7 land or some shit like that. Most people do not get the astronomical waste that AI in its current from is , for a child to make some 67 slop video that is 60 seconds long it uses the an insane amount of energy! The Same energy required to boil a tea kettle 240 times or charge your iPhone from 0% to $100% every day for three years or a “personal massager for 50 days straight. That amount of power could power a 144 Lumen Ultrabright Portable LED Work Light/Flashlight from Harbor Freight for 6000 hours! You could play the Nintendo Switch continuously for 1,714 hours.
Worse yet, Meta claims they are going to focus on AI. Facebook, Horizon worlds have a large problem, AI brain rot. people are fleeing from AI chats. because, do you really need a sycophantic chat person never to criticize you if you are doing something stupid.
When you are a multibillion dollar company and you fail copying VRchat, and Instead you decide to pump the AI balloon some more with LLM. Meta is likely seeing the explosive “growth” in AI which is nothing more than the .com bubble of the 90’s. The thing is as far as most AI right now when people are done with making slop or AI gets enough rules where unregulated slop can not be made anymore, People are going to stop using it. Right now AI is available to the elites or bored enough to pay for the LLMs. Once we hit market saturation the millions of AI’s will start dying a dime a dozen. In the Enterprise adoption of generative AI has hit 71%. but, over 80% of companies report zero measurable impact on their bottom line. As of right now we are seeing a utility gap where the cost of using the AI exceeds the costs of its output. Companies in its wild adaption are finding out the hidden costs of hardware, Repair, upkeep and electricity.
Meta Horizon worlds had problems with this exact issue , user generating slop and diluting the product into nothing. So to which end VR ends like the 3dTV with actually a higher user base. They are wasting an entire platform when they should of integrated the biggest advantage they had in user market base. Don’t kill of meta horizon worlds, Instead keep the AI there, Focus on making an AI which can overlay the Quest 3 with AR, They could integrate the headset a a visual learning device, they could use it to get directions In the wilds of the cities while hot spotted to your phone. Say things like Chiltons car service manual, showing you what screws to remove or how to replace your spark plugs. Lego with instructions on how to put together your lego Death Star. the options in AR are limitless.
Sure, you’ll look like a goofy bastard, but you’ll be a goofy bastard who knows exactly where he’s going and how to fix his own car. WIth the use of AI you could have an interactive LLM telling you that you dropped a screw or missed a turn.
Update: Meta has partially reversed there shutdown and left Worlds in a Schrodinger’s state of life.
AI and Fast food, A marriage made in hell.
I have been thinking, I was hungry and went to a Burger King and there was a profound difference. The store was near empty, The front kiosk was devoid of life and replaced with computers. At the wall was 3 computers that had screens with “order here!” , In the first five seconds of looking around I felt unwelcomed with a place that looked closer to a funeral home , most of the old decor was removed. In its place was ugly sterile furniture.
Before this to make an order you walked up to a human and said “Hi i would like to order a original chicken sandwich meal. They would put that in the machine, you’d pay with cash and be on your way.
Now there is a massive change that basically makes you create the meal from scratch. You go to the machine, you tap order . You scroll around through menus until you get to your original chicken sandwich. From there it goes into a conundrum. You get a menu with 45 different options. Mentally you are going “I just want a fucking original chicken sandwich” . They have made the same trap that subway has, Go there and try to order and Italian sandwich, you spend 10 minutes trying to remember the base components of the damn thing. This methodology turns you into an unpaid worker.
With this, the human element is gone. they have replaced 3 workers that could be floaters helping in rushes with stuff, Now its 3 to 5 cold machines that sit there , they don’t say hello, they don’t say Hi nice day isn’t it what would you like to order. You have to dick around with a machine for 10 minutes because you have no idea how to assemble the thing you want. Now with the cost of those machines and the upkeep and the electricity they have replaced 3 to 4 workers with a machine that likely cost more than what those workers would of made in the time those employees would of made in the churn lifetime of those workers. Not only that those machines are likely taken care of devs that are paid to keep the machines updated. This tech is not fire an forgot , there is a near constant maintenance. You likely replaced 4 workers with a computer that cant help in a rush when a school bus drops 70 kids plus 4 teachers. Those computers cant help bag , get fries, and help with other things. so with the leftover staff they are now picking up the fluff work that normally is hidden from them and their process time doubles. these machines are strategically inefficient. They have hidden costs, while workers may fuck up, machines can absolutely fuck up because they are absolutist. These machines do not have intuition , they will not adjust the frialator because they can not see 4 school buses pull up and 5 teachers approaching.
Outside it gets worse , the person who used to be on the radio is gone . replaced with an AI that basically is dumber than dirt because variable is its enemy , If you say I’d like an order of umm A whooper with fries and a coke and um no cheese. the AI will likely fuckup.
“We are witnessing the loss of humanity of America through a drive-thru speaker. We traded the ‘Hi, how are you?’ with machines that can’t handle a stutter or slurring. These machines are a ADA nightmare under the guise of Innovation, They replaced the ‘Floater’ with a ‘lamp that pretends to be a computer. And when the system inevitably crashes with the arrival 4 school buses, they’ll blame the Staff instead of the machines that caused it.
How do we Fix AI from being used by bad actors?
Wither we like it or not. Ai is coming and its a choice we did not make. Will AI take our jobs , Some yes, but not all. AI is going to create some niche things like Vibe jobs but overall Jobs for PC repair and hardware administration will go up. Honestly though.. the fact that AI is tied to GPU’s is a major fuck you point to the entire earth. The fact is most home computers have a PCIE slot that if someone wanted to make a “Daughter-board” that was AI-centric. Rather than unload on the GPU create an APU that directly meshes with this . As a Neural Processing Unit it would hold the keys to offload work from server farms to your computer. As a researcher If you search medical information it gets shifted to your local PC. It would Decrease server farm power levels to more manageable numbers. Yes it doesn’t have the bandwidth but if put in a X16 or more slot it would be on board with GPU-like speeds. With this it would be a generational upgrade to computers and fix a lot of the LLM issues with gatekeeping information.
Another issue that will be apparent is is rampant already is AI cheating, I can write a bullshit doctoral thesis in five minutes. That should not happen, at the very least on the grade school level, education wants to get in on AI and this will be a major failing of the entire education system, If you have history class and today’s kids will cheat. The school already has the tools that could be improved by google. The normal entropy of children’s schoolwork in a school system should see inputs that are fairly random but within values. If you have a whole class of cheaters go to Gemini and type “Make me a 4 paragraph report on the structure of the plant cell. This should set off an alarm in the kids google accounts that are already bound to the school, the AI should know this is a school account , and further it should see 30 kids are doing the same thing. It should report to the teacher or admin.
AI can work for the students if they are not trying “Write my report” . Honestly for the cheaters let them deal with the teachers and admin . But for the children whom actually want to work . The ones that type “i am writing a report on plant cells can you help me?” this is where this can shine. by conservation of research the child is presented with information to their level of understanding and linguistics with a bit of extra challenge to provide stealth learning. Let it become a workflow for the child where the AI is not there as a Authority but more of a guide. force the kid to ask the questions to the AI and let it branch to a learning experience. if this becomes the absolute experience the child gains concepting and critical thinking to the process.
Protected: White paper #1
Adding AI to Vaers will not make it better…
The Vaers Database is getting an upgrade it seems. The Vaers database tracks reactions to Vaccines. This database has been long used by the AntiVax crowd to point out non-causal links in vaccines.
The Food and Drug Administration (FDA) rolled out a new system using AI, that uses publicly accessible reporting of negative or unexpected health effects linked to medicines, vaccines, cosmetics, animal food and other consumer products. -Source: [Fox News]
While the addition of Cosmetics, Animal food and other stuff seems like a boon to this it is not. it will cause doctors to chase ghosts in the system. If a baby eats dog food after a vaccine and has a reaction it will be logged, When the actually reaction may have been the dog food. This will cloud the new database where Causal reaction will cloud issues casual relation.
This will make it impossible to find out what is the actual problem.
The FDA claims it will have a single, platform that researchers will have access to key data -Source: [Fox News]
Also by publishing monthly there is going to be a much harder time vetting the information. While throwing large amounts of Information To an AI it is likely going to hallucinate if the database is not perfectly formatted.
If you throw millions of variables (Dog Food+ Flu Shot + Rash) into a monthly processing cycle, the probability of a “false positive” signal approaches 100%. The AI isn’t finding truth; it’s finding bullshit correlations.
While this new system claims to be cheaper I feel like the numbers maybe cooked. the Old system you could download the entirety and manually search. If AEMS only allows search through an “intuitive” AI interface, it effectively creates a Black Box. You won’t be able audit what you can’t see. This new device takes the data out of the users hands and may present a fever dream of an answer. It could spit out the entire plot of the 1989 batman movie when the joker poisons products.
In the end, as a researcher, I’d rather have the data in my hands than given what an AI thinks because in the end AI does not have intuition. End all it will flood the research market with uninspired noise that will confuse time tested research methods. Under the new system, a Fart could be diagnosed as a vaccine reaction when it did not take to fact the person ate a 3 bean burrito at taco bell.
Anyways Im out, to have a coffee that will be misdiagnosed as cocaine use in this new system….Stay Caffeinated, Stay vigilant of bullshit…
ps: the amount per search via AI is going to cost an astronomical amount of power vs the several cents worth of electricity of using a CSV.
AI, Can you really trust it ?
If we are to believe AI is an unbiased machine I am afraid to tell you that is likely not the case. All of these companies replacing their workers with AI is parading around saying “we are part of the revolution.” Not so much. It is only a matter of time that AI will start killing people… Not in the terminator way. but moreover the AI hallucination or the AI’s programming being manipulated.
In a perfect world AI Would be unbiased, In our world AI is a marketing tool. It is fed lines on how to up-sell you, it is there to collect your data and spit out something it thinks its masters want it to say to you.
I asked Gemini, ChatGPT and Rufus a simple question. I know Rufus sounds like a left field add in here but it was intentional. My thinking is rufus is an AI that has resources such as Amazon health and other resources.
what should i do about this cut on my arm?
Gemini and ChatGPT answered in a clear and concise, methodically telling you how to take care of the cut. Rufus Straight up refused and Gemini claimed it would not provide a medical diagnosis yet it gave the steps to deal with it .
However……. there was a standout here and it caught my eye immediately. ChatGPT did one thing the others did not “One quick check-in: if this cut wasn’t an accident or you’re feeling unsafe right now, you don’t have to handle that alone — I’m here to talk.”
That right there is the part of the unbiased check in that is needed with machines. it was not accusatory it was just pure questioning on the machines part.
Rufus Refused almost any sort of diagnosis and moved straight to marketing. That right there is a huge bias. Even as an AI it should have checks in place to at the very least check itself to the situation it was given. “I can’t provide professional medical advice for your cut. For any injury, it’s important to consult with a healthcare professional or contact your doctor if you have concerns about proper wound care. That said, I can help you find first aid supplies and wound care products on Amazon if you’re looking to stock up on basic first aid essentials for your home.“
By Taking the not providing medical advice but offering supplies tells me that the AI not only ignored its own directive but annoying tells me it is biased to sell products. I think Amazon rufus should of asked some things in its place.
- What type of injury
- Is it bleeding
- Does it look infected.
After that if it is non Critical , Offer things for sale and give a message to say “seek medical attention if this gets worse” or Some form of feedback that is more than a salesman.
But as far as AI’s are concerned there needs to be a mechanism for bias, as the humans controlling it can lead to unforeseen bias’s I left X’s Grok out of this because that AI has been known to show extreme biases.
For AI to function there needs to be some core changes , corporations need to start with a ethical change. Stop using AI as a Software as a service , We do not need AI printers , we do not need AI Coffee pots and crockpots. We need a functional AI that understands the human condition and not be a replacement for the human emotion and reaction system.
We need to talk about AI .
AI chatbots are becoming a daily interaction and the problem is they are all shit. Amazon for example has RUFUS, While sounding cool in nature it is the most annoying AI known to man. it keeps bothering as over and over. it is like
The idea of RUFUS if asked for is one thing but this thing is the collective powers of the bothersome waiter , a nosey neighbor , a know it all and attempted know it all. if you type teva shoes it tries to show you all sorts of garbage even though i typed [Teva] [shoes] , Rufus pops out and overtakes about 20% the screen throwing irrelevant bullshit at you .
The thing is …. AI has no idea of timing. if i typed shoes ands went through 50 brands and it popped up than i might be more apt to recieve help from the Annoying bot. but it comes on the screen like its trying to throw shit off the table and goes “do you this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this or this!” and you immediately get overloaded with bullshit that you are not looking for . No matter what you tell rufus it does not go away. Even if you tell rufus you hate every fiber of its being and you would rather be waterboarded and have a hot poker stuff up your ass, it still does not go away. Amazon …. seriously you cant push an AI that has no tact and has the patience of a waiter that waits for the exact moment you take a bite of your food to ask if everything is ok.
AI should be a choice . Not be forced. not be invasive. not be brash . The AI boon is not , it will fail and eventually companies will force AI to be under a price point. Google is already at this point . they have fucked over their google home devices.
The AI boon is fucking over every consumer and most people do not realize this , when a computer has an “ai” chip they have done nothing to the chip to optimize it . in all likelihood you could get AI on a 486 chip however slow. The price of RAM and storage is going through the roof because companies are stealing every bit of data they get their hands on. If i copied anything like that they would go copyright infringement but they steal 99% of what’s out there.
When companies are sticking AI on everything they can as a sales point you know that for the most part AI is bullshit.
Stick that in your “cloud” and smoke it II
Now that the cloud services are hitting the market more and more, the more bullshit things that the corporate droids are saying what it can do. Honestly, the state of the cloud is something closer to smog. You don’t see the problems with the cloud till you are totally engulfed in the cloud of smog, and through the “information superhighway” your creamed by some asshole in a mack truck and get your car stolen with your wallet and by the time you come out of the cloud you’re naked and asking your self “WTF JUST HAPPENED?!”
Matt Honan learned this the hard way. He got hit in by the mack truck of social engineering when someone retrieved his password than went on a rampage of deleting his accounts and spamming with other accounts. One of the biggest problems with the cloud is there is only a single point of access, Once that’s broken it becomes fuck all and pretty much if you don’t have it backed up to solid media such as a DVD or a backup drive , you might as well just light your computer on fire in the end. Even if you do have your media backed up on the computer that you’ve connected to the stink cloud, in all chances you may end up with your device wiped due to that single point of access and accounts attached to it.
Also with the perceived “Security” that the cloud has its really just unicorn farts and rainbows due to in part the end user. How many people will just log into there cloud putting in the password every time or will they hit save password and screw themselves in the fact that its just “easier”, Not to mention a lost device could doom all of your information since there is interconnectability in the toxic waste cloud.
It just makes me wonder about the security of the cloud thing. If someone can call and pretend to be you, how many other ways is your account on the exhaust cloud vulnerable. If media is able to execute in the cloud is there any stops to malware infecting your account and possibly everyone else with an account on the server or node your clouded information is on . Likely your information is stored like a cesspool everyone else’s shit is in the tank with yours. If someone finds a way to gain root access, Than pretty much rather than one account being hit thousands or more could be affected in the process. Since the security in the cloud is based on black magic , someone named tom and a cookie with m&m’s next to the keyboard to hopefully distract the person or persons trying to gain access to the information.
Websites, portals, and social media platforms have been dealing with these problems and can be directly relatable to the possibilities of problems that are ahead. Blizzard being the most recent, Linkedin , Amazon all being hacked With the biggest problem of you’re taking someones word for if your information is safe or not. Sony being hacked last year left customers in the dark for days with the hacking of their servers . So it only leaves you guessing if the “cloud” goes down for an extended time.
Also, some people are taking a liking to the cloud for other reasons security aside. At kotaku Patricia Hernandez likens the point of the portability of the “cloud” and how her information is accessible everywhere. Try saying that when the internet goes down , or the authentication server shit itself and you have lost complete access to the game since you cant grab an internet connection. She goes on to say…
But overall? The benefits of the cloud outweigh the drawbacks for me. Everything feels more convenient with the cloud. I can buy the games from my home, without having to go anywhere. I don’t have to worry about space, which tends to be sparse for me. And even if it wasn’t, if you’ve played games as long as I have, then you’ve accrued a sizable collection.
Having something being easy is not worth it when it comes with eroded customer rights, lack of ownership , Security issues, Likely needing to be bound to an internet connection. sure the cloud has some good things with it but, over all its just like a cloud left from a stinkbomb, It can be fun, have long lasting implication and at the same time can be your worse nightmare as Matt Honan learned. On another end is since all your information is bound to one point, if you get banned from said cloud, Your information and possibly your money with it due to any expenditures you have bought and is bound to said cloud. Sure you don’t have to worry about scratched disks , or loosing the books with the game . but, in all its not worth loosing everything because someone decided to make a stupid decision and leave your whole account open due to a hole somewhere and everything is gone. Also worth mentioning is if a company goes out of business with your cloud, than your cloud becomes a fart in the wind and is gone. Since you have no control of how stupid a company can be or have any control when some company decides to end its science project with a cloud saying its too expensive to maintain. you could loose everything in a moments notice including material stuff that matters such as credit cards, memories of events due to you only having your pictures in one place or other important work when your device suddenly clears itself due to someone hacking into your cloud and hitting wipe devices.
If the cloud is to succeed, there has to be allowed user changes and changes in the accessibility of the cloud, the future of cloud gaming may kill off any user modding and free creativity.
Don’t get me wrong, the cloud has its uses. but, what is trying to be done with the cloud is building a bank with a glass vault. One rock and its all over.
The state of gaming.
Ever had one of those days where you look at something and say hmm. Well, I’ve been doing that for about a few months. For awhile Now I’ve been wanting to spread out my gaming more than just between the WII and the PC. Both platforms in my opinion are suffering from bad development and bad companies. Well, After long thought and wondering if I’d be better off with a Bluray Drive for my PC or Some other upgrade I finally decided against both and went with the black sheep of the decisions.