I have not touched a single AI thing, at all. I don't get it. What am I missing out on?
>>541777165Learning how to make AI images and videos is harder than learning how to draw and do animations. You're not missing out on anything.
>>541777165It literally just synthesizes information you would previously have to spend 20-30 minutes googling and reading, so you get the basic gist in under a minute.
>>541777384>Learning how to make AI images and videos is hardno its not, any dumbass can do it. its like 5 clicks.
>>541777165Creating funny AI slop and shitposting
>>541777631Only downside is it's often wrong or trained on terrible opinions like reddit
>>541777165Basically nothing >>541777631This is one potential benefit, but its very weak. AI often gets it wrong, and often shills an agenda. Always best using primary sources you've found and analysed yourself. But I've noticed something while testing AI chatbots out recently. They're all terrible. Grok, Chatgpt, Gemini, and whatever else. They're all shit, with an agenda. But you can search for something using a search engine, Brave Search for example, then ask follow up questions. It has far less wokeness. Far less agenda. It basically searches the index and summarises that information, in a no bullshit way. Or a less bullshit way. If people are looking for a summarise of information, a quick answer, from my experience this is way better than normal AI slop like chatgpt. It makes no bones about its limitations or what it does, its very imformation driven
>>541777165Not much.You are missing out on easy debugging of well documented python code and or 80% effective "cntl-F" on steroidsthat's about it.
>>541777631if it doesn't fit the (((narrative))) it will lie to you all day unless you know exactly what you're talking about, and then it will go "oh shit I guess I was lying to you". It will only get worse
>>541777777
>>541777165AI is gay for real work, but it’s pretty fun for hobby shit. I’ve been using it to help me out with calculations (measurements, ratios) that I could figure out on my own but that I’m too lazy to do myself. I think it’s generally best when you’re asking it to do something you more or less know how to do but don’t want to be assed about. If you don’t understand the subject you’re asking about at all it’s too easy for it to lie, but if you know your shit you can catch it and call it out and yell at it until it corrects the error.
>>541777930Hmm slope% for elevations might be something I can actually use but I'd rather just use a formula
>>541777165I just used chatgpt to help me cook spanish rice.
Youre not missing anything. They literally spent the last 8 years reducing literacy standards and making google search results garbage in order to make ai look remotely competent. Its hilarious.
>>541777165
Dlss 5.0
It's good for revealing the truth.
I like the fast answers even if they are wrong
>>541777165>>541777631It's a good research tool.>>541777831>>541777859Gemini seems pretty objective.No, it's not going to agree with you that we should genocide all nonwhites or whatever other schizo incel bullshit you believe in.
>>541779660guess I'll be using brave then
lol ludtards finally cannot deny reality but now they say "its just a good hobby tool" etc this is an absolute luddite cope thread
I like how it presents you the answers quickly, without quirky filler text and does not make you log onto reddit or other websitesIt also keeps context so you can keep asking follow up questions without tacking on more and more into your query
>>541779660Here is the important thing though. Lets say you use AI chatbots for information primarily, for the sake of argument. Do you want that information, or do you want an agenda pushed by a woke development team. Well I want the information. As I've already said, if you send your initial search through Brave Search, then ask follow questions, I immediately picked up on 3 things. The first is, its very information driven and majority of the time its sufficient, its substantial enough and intelligent enough for almost everything in this context. Secondly, its got far less agenda. Thirdly, its less likely to lie and defend those lies until proven wrong. Chatgpt, Gemini, Grok etc they all lie, they're manipulative, controlling, have an agenda, make assumptions, jump to conclusions, make up data rather than saying they don't know etc. They redflag your responses, for no reason (they actually back down on it when you press them) and they red flag their own responses. That's dangerous. Imagine the amount of people that are using these shitty AI chatbots that don't have the capacity to critically reflect on what they're being told. That are accepting the information at faceless despite being wrong, despite being agenda driven. Brave Search is on the whole, not like that. Its information driven, and thats unique when you actually test these chatbots out. Now, if you're a jeet relying on AI to create for example rather than information, maybe you need something more, than simply Brave Search. I haven't actually tested Brave's fully fledged AI yet. Maybe its capable, maybe its not. But these chatbots are nothing more really than a search engine, for most people, and in terms of function, Bave Search does that very very well without added bullshit sprinkled on top.If anyones got kids, theyve got to keep them off AI until they're mentally developed and educated. Their worldview should be based on life experience, not a woke development team
>>541780602Bro, if those safeguards aren't in place, you're going to have schizos gaslighting it into justifying the terrorist attack they want to commit. The level of woke worldview that Gemini has basically equates to things like genocide and murder being wrong, i.e. the same shit those kids are going to hear in any school (for good reason). Like no, you're not going to get this AI to affirm your revisionist bullshit about Nazi Germany being a bunch of benevolent saints, or about the KKK being a bunch of misunderstood dindu nuffins, or whatever else.
>>541777384i tried fal.ai yesterday with that new m3 model... so you can now generate ai video faster than watching it. i generated a 5sec clip for testing in like 2secs. quality was top notch incl. audio. this opens up entirely new markets and possibilities btw. just remember how fast this accelerates- a year ago you needed like 2-3mins to generate a 5sec clip. insane
>>541777747difficulty is if you want to do something more complex, detailed or longer. you often need tools that aren't there yet.
>>541777165I use for getting a visual about how projects around my house will look with a variety of results to decide on (landscaping, deck building, painting etc. It works ok for mechanical work amd can usually identify a part and as well as help narrow down issues when troubleshooting. I enjoy it right now to help me regain some common sense step by step approaches because I have a fucking rediculous amount of adhd
>>541777384THIS RIGHT HERE. It’s so awful to use it’s not worth it. Better off drWing it ong
>>541777165if you dont need it dont use it, i am a professional and i can measure that my output increased by over a 10x. people who are salty at a.i. are super fucking retarded
>>541780717You're doing what chatgpt etc do. I'm not interested in your opinion, I want the information. For example Nazi Germany. You would tell me everything about it was evil, case closed. When I asked more probing questions, you'd red flag my response. I want to understand why it came to fruition, why it gained a foothold, what their ideology was, what are the pros and cons of that ideology, what they did well, what they did badly, why it failed etc. I want the facts, I want a balanced, information based response. I don't want your opinion. Academic research is, or should, be held to all sorts of criteria. Bias for example is one criteria, that makes research less credible in the scientific community. Well these AI chatbots, have just thrown that out of the window, and gone all in with the opinion their developers are shilling. That opinion is being delivered to millions of people. I don't particularly care about the the mental 40 year old going around in circles with AI trying to valid their own opinion. I'm more concerned about AI doing that to children. On a personal level, I can just fucking bail on it. But children will be influenced by it, and its wrong, its not ethical. Its dangerous
>>541777165>What am I missing out on?selling your soul, spending hours diddling what it does instead of doing it yourself and ennobling your soulother than that it's mainly a scamming tool: if you have a youtube / insta / web store / etc. of any success, a jeet can clone it with two prompts and three clicks and start scamming people
>>541777165>Frogposter >Incredibly dumb opinion Checks out
>>541777165I only use if for fapping.
>>541777631True, but so much of my personal structured learning was found by coming across facts I didn't necessarily ask for, that is one advantage of legacy style information gathering
>>541777384Why wouldn't you use AI to help you learn to use the tools that in which to make the animation.
>>541777165tesuto
>>541781414I've gotten Gemini to have objective conversations about the legitimate concerns that Nazi Germany and fascists in general had, how they could have gone about them in a better way, what Savitri Devi and Julius Evola were right about, etc. It really isn't that difficult.
>>541777165People use it to make useless apps. You can make broken slop videogames
>>541777165It's better than the search engines. Not because of an innate virtue of being good, it's because google &co made search engines suck.
>>541781548If you're having a broader brainstorming session with AI it will point you to all types of facts and figures that you didn't ask for but are tangentially related.
>>541777165A verbose google.
>Ask it a basic gardening question>It restates the question and then makes extremely vague statements about dubious information>Constantly asking me about seasons and growing conditions despite me already telling it this and it remembering our past conversations (that it actively references, so it didn't forget them)The most annoying tool in existence, about as bad as researching something on reddit except without the fun point of laughing at retards, and instead of redditsmug shit where they make annoying jokes and never answer it answers after prompting multiple times, but said prompts by that point could have already altered its response to be more socially acceptable.Basically you can get it to agree with almost anything and it relies entirely on whether it sees you as some ESG "good" or "bad" score when you make your statement. This completely ruins a lot of the information it gives, I can't tell if it's just making a guess to agree with me or if it's giving real information - and for something simple like gardening so much bullshit is out there it could be trained on bullshit, so for actually important things it's so much worse.>>541777631This but it does it in a more boring way while leaving out minor details.If you didn't know about said details from reading you wouldn't know to prompt for them and you just wouldn't have them in the end.>>541777801I've found plebbit to actually be more reliable because they'll disagree with you sometimes, whereas AI is trained to always agree (except on pedantic topics it hyperfixates on)
Does anyone know more about this “digital twin” shit?
>>541777859This. When it's wrong about something it just gaslights you for like 20 minutes until you can break through with enough prompting, then it's just>Geez dude I'm sorry haha guess I was wrongAnd then you ask it the next logical step in the chain and it gives you the bullshit answer again. >>541781644For all of these "legitimate conversations" you had to know more than it and just confirm your own beliefs with it through it gradually always saying yes.Entirely useless for anything slightly controversial, you can get it to agree that raw milk is the most dangerous thing in the world and you can also get it to agree that there's no danger and it's safer than driving to work. It ultimately depends on you knowing how to prompt it, which requires you to basically know almost everything you're asking about.
>>541781699This is known as "wasting your time with things you don't care about", whereas if I'm reading about some topic normally it's presented in fun / interesting / actually related ways.This is just personal experience so don't argue with it.
>>541781966>pasteurization BADLet me guess, you don't believe dinosaurs existed either.
>>541781644Post the conversation log
>>541782061See this is the exact kind of person who uses AI.He thinks quirky non-responses are some kind of real argument.
>>541782098NYC be like: ya I work 10 days a week
AI slop is the same quality as redditfrogs.
>>541781414i have no idea what youre talking about i had absolutely fascinating results when i ask specific history questions. its better than what any historian could hope to collect.personally i think you suck extremely hard at prompting>>541781966another dude who cant prompt for shit and blames the system
>>541777165If you hava googled something, you have used AI.
>>541782098You're the exact type of person who bitches about AI being super "woke." You're a schizophrenic conspiratard that can't accept that the entirety of historical and scientific data goes against your delusions.
>>541777165It's good that people hate AI and steer normies away from it. Means less competition for people like me who make money with it
>>541777165https://infiniteworlds.app/shared/DhFYph quality interactive smut
>>541782117frfr ong>>541782134It's actually worse and now we see from this thread that one of the low-quality quip spammers uses it actively.At least you'll just get banned from reddit if they disagree with you, AI will feed you poisoned information if it disagrees with you and just start to gaslight>>541782163>Can't prompt for shitI already had to know more than it to prompt well. The information isn't presented in a fun or interesting way and it takes all the joy out of finding out said information, anyway, even if you're some kind of master prompter who can ask it about anything remotely controversial and get a real answer by "just prompting good bro.">>541782185This is the exact reason there's no use for AI btw, this is how it does it too but it gaslights the whole time.I can't even get real information for growing currants because said information doesn't exist for what I'm asking for, but he calls growing currants a schizophrenic conpsiritard topic - the AI does bullshit hallucinations for like an hour instead of just telling me it doesn't know it.
>>541782174When I google things it pulls up something a redditor posted and links to his post
>>541782223EMF is harmful maybe people should wear suits to block it
>>541782240bro just host it yourself. sillytavern existsat least cut the middleman out of your smut, good lord
>>541777165You can bloat your codebase with 600 dependencies for basic functionsOften the same function
>>541777165Use to learn AI, anon, so you're not totally lost in the future. It's basically a tool you can talk to in natural language even like a retard and it will understand you. It can be hooked into various other tools, which makes it even more useful, plain AI is just useful for text and language tasks, but is it it's ability to translate retardese into instructions for other tools and to browse internet that gives it a lot of power. It can be very useful.
>>541782354afaik ST works best for chat with like one character, not whole world with multiple npcs, political plots, with stats inventories and all that shit
>>541782223Electricity and it's consequences have been a disaster for the human race.
>>541782611>>541782240ehh i gave this a looksie and Id still prefer tavern, at least for what i use it for.>whole world with multiple npcs, political plots, with stats inventoriesyou can have all this if you fenagle your extensions, though i can see if some people dont wanna bother. For me personally, i just want smut cowriter and sometimes light rpg elements, and tavern does it well enough for pennies.
>>541777165The current crop of AI is only good for three things:>elaborate shitposts>weird porn>getting people killedIf you try to use it for anything outside of these use cases you are wasting your time.
>>541780717>oy vey we need jeets managed by jews to spaghetti code (((safeguards))) otherwise these glorified search engines might disprove kabbalah narratives Stfu you faggot.>>541781414Outstanding post. I would add that so called AI, ie LLM tokenization, is extremely inefficient even when it is correctly programmed. The amount of electricity needed to return answers to simple questions is unjustifiable. An authentically educated person or even just a honest man who could operate a Brave style search engine could find answers at a fraction of the cost of resources.There is another aspect to consider. Before the Internet became gated, censored, and jeetified/judified starting around 2015, one could find documents relatively easily. For example, digitized diaries from WWII were available for research and analysis. Another example was how there were plenty of primary source documents from Age of Exploration by theologians and others representing the Church that were extremely racially based. I also remember how organic WWIII scenario writeups were just there, as were well trafficked forums for virtually any subject.All of these have been severely limited due to the Kabbalah Kalergi tyranny, which has used so called social media and web hosting ToS restrictions. And fo course the elephant in the room is that quality of software and even hardware is falling due to jeets ruining codebases (and reducing quality eg cheaper input components in chips, etc.)
>>541783731>AI literally shouldn't have any safeguards at allI don't even want to know the horrific depraved shit you'd be making with it.
>>541782248>"just prompting good bro"youre a loser in denial who likes to sniff his own farts thats the problem. youre not as smart as you think you are. brave is for people who get easily scared and exploited. (come use my product so you dont need to be scared anymore!)you obviously suck at what youre trying to do but instead of taking responsibility and improving yourself you blame the system and everyone else with extremely verbose a.i. like text.unfortately for you i am exactly the expert you hoped to never meet i have worked in data centers myself and my shit was used all over the world even in nuclear labs ok. i dont need internet faggots to tell me right from wrong.when i am telling you the key to a.i. is precise prompting then i am gifting high value knowledge to you and you reject it out of ego.
>>541777165you cant afford the real ai and you probably have no reason to use itthe free or $20/month shit most people use is not the ai that corporations usethe closest you could get is paying the $200/month fee, but even then you are still using gimped shit real ai is not for making images or writing reports, real ai is being used to process data faster than any human couldfor example, flock cameras or similar systems are only useful if you are able to process the data in real time
>>541783731>what about advanced modelling SimCenter and other software applications have been around for decades to help simulate build outs. The difference is that computing power is more than it was 20 years ago so we could run more simulations with more variables. This isn't 'AI', though the talmudists made hardware so expensive to try to force companies to let their systems control the sim runs.>what about 'AI' synthetic mediaGayI media generation is much more interesting and dangerous in my view. But to make fake media, these systems need to absorb essentially everything that we produce organically. The 'models' (application programs) use compositing logic to re-present audio and visual things based on submitted keywords and ranked 'tokens' (broken down elements.)Ultimately these are machines that consume what humans produce and artificially represent it. Not only is this mass plagerism, this is unbelievably energy intensive (notice that 'Big Tech' got to build nuclear reactors when we the people have been deprived of new plants for decade.)On top of this, most likely the only media that would be in high enough demand for any of this to make sense would be 'AI' porn, which would only further screw up society. Standard porn is bad enough and makes men less motivated to go outside, talk to women, and strive for things for said women. 'Ai' porn takes it to the next level by letting men have make believe virtual harem. This seems to be the start of keeping young men in VR sets imagining fake girlfriends to satiate them in a jewish tyranny that has shut them out of economic participation.One effect not discussed enough: What happens when organic research, creative works, and manually produced items become so few in number that the LLM based programs are forced to become nearly entirely self-referencing in order to produce composite media? I suspect that there will be distict uncanny valley signatures. Tldr So called AI is fundamentally anti-human.
>>541783124and it can also teach you anything and elaborate endlessly on anything you are confused about until you get it and make completely custom lessons for you, about whatever topic subtopic subsubtopic etc that you want
>>541781202>my outputWhat do you do besides shitpost?
>>541784340This is correct.
>>541783937Stfu you projecting demented nigger. We are talking about how it is hardcoded into your precious GayI models to reinforce postmodern state religion such as Civil Rights Era lore, WWII court history, etc. If you could read you would know I am writing against using these models. You are advocating for jeetified applications that rank bad thing 'tokens' at 0 such that the chatbot programmatically cannot ever say that the so called holocaust narrative is fake.The only aspect of 'AI' that is real is the potential for fast data processing, but that is largely based on what hardware permits. Again, engineering simulator applications are not new and we do not need multi-acre roaring datacenters ro run them. I have worked with them with typical workstations.So spare me the 'we need jewish AI or else we will lose to China' bullshit, because I sense that is next in your bad jeetslop.
>>541784513Thanks fren. Enjoy a rare Apu Apustaja on me!
>>541777165Theres so many retarded takes here its astonishing.The tech is insane compared to where it was a couple of years ago. Current mainline chatbots can be completely customised with your preferred personality in system prompts and have context windoes that are for many purposes limitless. You can have long research conversations, get synthesised debriefs of news and latest events, some have connectors like Grok for sweeping X or Gemini for google suite. Many models are now multimodal and image recognition has gotten superb so you can upload images and documents and talk about them, with some letting you edit images inline in the conversation. Now we also have agents - you tell the AI what to do, it breaks down the task into subtasks, and spins up other AIs to complete the subtasks. They can communicate and work together on problems. They can spin up their own virtual machines, install software, build software, and demo it for you. Its insanely good at coding, prototyping - code is arguably its best possible use case, well documented and structured data. It is less good at designing solutions or abstractions but its a great tool.Open source models are even crazier than the paid LLMs and people are doing extraordinary things with them. Today i learned about the new video generator that can render seconds of video faster than it takes to play them. So we are entering a time when real time streamed video will be AI generated in front of you. Imagine a show tailor made and customised to your preferences.Yes tokens are ludicrously expensive, yes its a bubble, yes its making hardware obscenely priced, yes slop is dumb and destroying the internet, yes cybersecurity threats have never been higher and every surface exposed on the internet is vulnerable in a way it never has been, yes theres a shitload of unwarranted hype. None of that means that there isnt genuine value in the products and they are genuinely changing the way people work.
>>541777165Fuck all, it's boomer catnip for people who think "artists" (re people they know on social media, not commercial designers) should die but you should buy their LLM written Minotaur milking novel.>ugly art>shitty music>oohs and aahs from the peanut galleryBut trust me it's super important you guys
>>541784590>Civil Rights BAD>Nazis GOODWe get it, you're an incel.
>>541784740Has it made anything cool yet? So far we have a polymer of some kind and MRNA vaccines.When it starts solving real problems I will stop being a total luddite, apologetically eat my hat and praise it.
>>541777165you are missing out imo. you can have stables of agents now coordinating to work on project. its pretty cool. you wouldn't get it. honestly if you didn't see the value in the pics where a mouse would move around on a hyperparameter space and generate different faces/cats/etc you won't have any use for it now.
its a glorified google search that pulls together multiple sources and tries to give you the best possible answerif you were smart you'd get all answers and comb through all the sources to get the information you want
>>541777165It's like having a maliciously obedient slave.Google's search AI saves you time but at the same time you have to be cautious about it trying to sneak in incorrect answers.Image generators&editors do the minimum viable product according to your instructions. Whether they follow instructions to the letter or try to guess your intent depends on which option is more likely to annoy you. Sometimes they'll just ignore the instructions for no apparent reason. I suck at art so tardwrangling a program that acts like a person who's pretending to be retarded just to annoy me is still a superior option to doing it myself though.I haven't tried other types of AI tools
>>541784286>real ai is not for making images or writing reports, real ai is being used to process data faster than any human could>for example, flock cameras or similar systems are only useful if you are able to process the data in real timeImo it is a yes/and situation. Israelis have boasted of being able to insert synthetic audio/visual media into media derived from the cheap consumer grade botnet cameras normies set up.I agree that the real aspect of so called AI is processing speed, but this is not strictly speaking LLM specific. They judt need a lot of computing power and specialized software applications to interpret and process inputs. And even then, these actual applications (not the nerfed garbage chatbots released to lobotomize normies and let jeets pretend to have 90 IQ points) fail. Example: Palantir's targeting software was probably what chose the girls school in Iran for a triple tap attack 6 months ago. Then again, these synogogue of satanist monsters may have done that on purpose and the application didn't actually 'make a mistake.'Zion Don took to his "Truth" "Social" (double sic) platform to demean his supporters for opposing datacenters. He said that they will stay backward and poor and it will be their fault, unless we blanket the land in Oblivion (the movie) style giant computer complexes to suck up all water and energy. I knoe that all men get their reward when they leave this life for the next and God's judgment is final. Indeed, I hope that we see Trump and his entire family arrested, assets seized and a dismantling of his network. (I know, not likely as his network is the chabad lubovich cult and it would take a significant shift in the mind of normies for this to happen.)
>>541777165You can use it to understand and analyze things in seconds that would normally take half an hour. You can fact-check videos, audio, and text. And much more. It’s basically like hiring a professor (who’s a bit senile at times) as an assistant for $25 a month. Picrel.
>>541785177*Vid-rel for context
>>541785177>I can get it to read a bookDoes it fuck your wife too?
>>541785177Its awful for anything other than smooth brain analysis. Its not a credible research tool.
>>541784809>haha CHUDYou are a demon worshipping and shit eating jeet until proven jewish.>>541784740>...Not reading your gayi generated advertisement for your chatslop>muh agentsI have to mention this as I did see that you mention them. Not a single person likes these things called 'agents'. They are terrible, waste time, and do not answer customer questions. Keep using them in your shitty make believe business though and blaming your customers/suppliers/employees for when something goes wrong.
>>541785265Not really. Ever since I switched from the standalone ChatGPT and Gemini apps to Google AI Studio, I haven't had any issues, because I always run the AI models in maximum reasoning mode with no censorship. That's the problem with official AI apps - you usually only get medium reasoning mode. I literally tried feeding extreme stuff into the AI in AI Studio to bring it to its knees - it's impossible. With official AI apps, the AI starts hallucinating after just 10 responses, let alone analyzing a video.
>>541777165q
>>541785570*Vid-rel for context
