Garbage data. Reddit, Twitter, Facebook, ETC.... trained on the best data we have online... Just pull up a sewage truck and pump that into your data center to train on.
Economy. AI will take everyone's job! Yeah! Who will pay for AI?!?!?!
AI will take all the power!!!!!! You thought home prices were high! Wait till you see what we do with your power bill. AI will 100X your power bill!
Thank Elon and Sam Altman for total destruction of our society!
No consistency either. If a calculator gave you a different answer each time you asked the same Q, you would throw it into the trash. And not even merely divergent answers that at least feign a level of plausibility and correctness, but brainless hysterics.
When AI is gobbling up nonsense from other AIs into its LLM data sets, there is going to be crazy and escalating levels of unreality. The 1st gen ones had the benefit of consuming purely human datasets, like a pristine world that had never been mined before - easy and viable value extractions.
But what happens when the internet is 99% white-noise digital junk and very little actual data?
I dont see any reason why the quantity of email thats spam, proportion of phone calls that are scammers, proportions of twitter that are bots, proportion of the websites that search engines will never index as potential results etc wouldn't follow exactly how AI is going right now. The same data-trust problem repeated over and over in each new medium and niche. And forming a viable chain of trust over the top of the hostile and unrelenting flow of nonsense is the answer for each case.
When AI is gobbling up nonsense from other AIs into its LLM data sets ...
And therein lies the problem -- and will ALWAYS be the problem.
WHO decides WHICH datasets are "correct?"
No computer can ever do that, which means it will ALWAYS come back to the FACT that a computer will be PROGRAMMED by a human -- who may or may not be honest or honorable, and will NEVER have enough knowledge to know which datasets should be accepted or rejected.
The AI will just spit out what some human told it to spit out, thus making "artifical intelligence" an oxymoron.
Human intelligence is enough of a problem already, and artificial intelligence is an impossibility.
Moreover, with scumbags like Sam Altman using their unjustly inflated ego to lie about what is really possible, we get a "tyranny of the 'expert'" problem -- just like the liars during the Covid scam.
Yes, because that is the fundamental mindset of someone like Sam Altman.
He is incapable of NOT thinking about how to control others -- just like any other standard issue libtard out there.
Their minds do not allow the concept of a free people making decisions that the libtard would not make, thus they think it must be "wrong" and needs to be "corrected" (i.e. re-educated).
Tyranny mindset. That's what these people are all about, and we should start calling it that to drive home the point.
I think Altman has been associated with the philosophical/ social movement, “Effective Altruism”🤔
If you can consider what he’s heavily involved in, OpenAI, Oklo inc. (previous chairman), Tools for Humanity (co-founder), Worldcoin (co-founder), Apollo Projects, Y Combinator and Loopt AND you know what those things are about you can see why he’s being associated with Effective Altruism.
But,,,, he’s a major backer/investor in Reddit and depending on your views on why Reddit is the way it is you might surmise he’s the ultimate control wizard! 🤷♂️
I just know that Elon Musk was interviewed about a variety of topics, and the subject of AI came up. He said he talked about it at length with Altman, and asked about the down side of the harm that could be done.
Altman was oblivious and uninterested in talking about that.
Ever considered humans have much more to give than just an 8 to 5 - - -?
Why have a spirit/soul whatever if you are destined to keep your head down for the majority of the day and work?
If you’re here you’ve broke free of a lot of what binds us as humans. But,,, it seems to me that the ultimate bind is how we have been sold the idea that we are meant too/better off if we work our ass off our whole life and then die.
Here’s the rub,,,, I LOVE working, but I don’t love working for others and having very little control over when my free time happens. I have no doubt that if I had more control over my time I would accomplish more “meaningful” things that would raise my consciousness and possibly that of others, ai seems to be the fix for that..
Anything you can reduce to strings of data and patterns can be used to train AI. Its not that it has no scope, but the scope is umlimited.
No clear definition of problem/solution.
Just like the previous point - the beauty of AI is that the number of different problems it can be used to solve is limitless. Anyone trying to come up with a "definiton of problem/solution" is forcing themselves into a box.
Garbage data.
AI is "garbage in, garbage out" technology. What differentiates different AIs is the quality of the training data. This is another beauty of this technology. There will always be better and better, even open source, AI models because anyone can train a model with their own data. Power to the people.
AI will take everyone's job! Yeah! Who will pay for AI?!?!?!
"Robots will take everyone's job"
"Internet will take everyone's job"
Both of these prophecies were partially true, partially false. A lot of jobs will become redundant. Especially the low end coders and you absolutely want these jobs to become redundant. But at the same time, it will expand what we as humans can achieve and create newer, higher level jobs that focus more on human thinking, intelligence, creativity.
AI will take all the power!!!!!!
Perfect forcing factor to unveil real energy solutions - be it safe nuclear or the so called Tesla energy.
It feels like you have never worked with software. The first step in creating a project that fails is to not clearly define the scope of work and to not limit requirements coming in.
We could give AI good clean data... where does that exist?
People need work and income. That is how our economy currently works. If AI can do everything why have any humans? Humans also need starter jobs to learn to work.
Power you are assuming the best case scenario. That the government will release the new (old researched in government lab) tech for this problem. I have a bridge to sell you... The bridge was in London at one point in time...
It feels like you have never worked with software. The first step in creating a project that fails is to not clearly define the scope of work and to not limit requirements coming in.
I think your perspective is coming from being a software contractor who takes requirements from a client and delivers working software.
Eventually you will get to see bigger picture of how not just a small piece of software is developed, but technology is created. Every single company that I worked in started off with a clearly defined problem, but somewhere along the way realised that their focus should be elsewhere and they pivoted. This is a standard lifecycle of not just startups but technology itself.
AI didn't pop out of nowhere. Precursor to AI was the field of Artificial Neural Networks (ANN), that started as early as 1950s. They had a clearly defined goal - to mimic the way neurons work in the brain and they had a well defined problem: Pattern recognition and classification.
When ANNs hit into the limitations of computational power, it was largely forgotten as a field and people moved on to more exciting areas. Deep learning algorithms - foundation of today's AI models - were just extremely complex versions of ANNs with much more optimized algorithms and the problem definition that can be solved by AI pivoted from pattern recognition to pattern generation.
All of today's AI models are nothing but pattern generators. You feed them billions of data patterns, and then you feed it a new pattern and it will extend it by generating the continuation of that pattern.
When you ask a question to Grok, all grok is doing is taking that string you typed and running it through its internal network and extending that string into something that, as a human, you perceive it to be an "answer" - but its just a string of letters.
So yeah, AI has a clearly defined function - pattern generation and extension - but what problems you can solve with this is unlimited.
To understand this concept that applies to all new technology, you can compare it with crypto space. When blockchains were originally created it was solving a well defined problem of decentralized consensus of financial transactions with no double-spending. Thats exactly what Bitcoin protocol implemented.
However, the problems you can solve with this is unlimited. Started off with currency, extended into programmable contracts, NFTs, and pretty much anything you could imagine - now reaching the problem of making liquidity available to people without banks as middlemen.
When we compared where we are now with where it started from, you cannot help but be amazed. And its the same with AI.
People need work and income
I guess you missed the point I was making, explaining how - just like with Robotics and Internet - while some jobs were lost, a lot more were made, and in the end we still have jobs for everyone - from newbies to veterans. However the jobs that newbies today do are different from the newbies did 50 years ago. And it will be the same with AI - because, and its important to keep in mind, AI is not a thinking entity. Its simply a pattern generator. We humans are the thinkers and we will always be that.
AI is not a thinking entity. Its simply a pattern generator. We humans are the thinkers and we will always be that.
One thing I’ve learned from you guys is to never think in absolutes. That’s not to say there’s not absolutes that exist, but from my experience the moment I think something is set in stone is the moment it begins to change🤷♂️
*Do you believe the human mind creates reality? Literally,,, creates reality?? Are we only limited by our understandings??? If you believe that and accept it as truth then is it really that crazy to “”think”” that Ai will develop a level of creativity that matches/surpasses human ability?
*”If” our brains have the ability to receive/tap into the universal consciousness for answers/creativity (thinking in terms of frequency/radio waves) than maybe we can create something capable of this too!
I believe humans have a soul which connects if to the Universe or God. This is what differentiate is from even the most powerful AI. They can be smarter than is in many areas, sure but we'll never possess our instincts, insights, and spirituality that has guided us as humanity even through all the deception of Satan.
It entirely depends on what your worldview about God is. If you believe God is a supernatural being as conventionally accepted in religious circles, then Satan is an angel who rebelled.
If you believe God is extra terrestrial beings who created our DNA and populated us here, then Satan is a ET who wants to treat us like cattle.
If you are an atheist - then God and Satan does not exist.
Its not that it has no scope, but the scope is umlimited.
False.
While in THEORY the scope might seem limitless, it will in reality ALWAYS have limits.
Did we go to the Moon?
How would an AI computer answer this historical question?
Would it draw from an "approved database" of information?
Would it accept data from multiple databases that conflict with each other? How would it resolve the conflict? WOULD it resolve the conflict?
Humans cannot agree on how to resolve this and similar conflicts, so how would a computer that processes 1's and 0's do it?
If by "AI" we are ONLY talking about computers that can process data faster or more efficiently, then sure, it has its use. A computer programmer can write code much faster with AI -- and some companies are firing coders because they don't need them to do as much as before.
There will be some uses, for sure.
BUT ... human intelligence is all about (a) taking concrete ideas, then (b) imagining some abstract concept that can utilize those concrete ideas, then (c) applying that abstract concept back to existing and/or new concrete ideas.
This is what IQ tests measure: the ability to apply abstract concepts to concrete data.
Computers can ONLY process concrete data. They can also do it much faster and more efficiently than humans. BUT they CANNOT come up with NEW abstract ideas on their own. THAT requires human input.
Thus, they are not "intelligent," in a human sense; therefore, the term "artificial intelligence" is a LIE.
This is the crux of the problem, because the tyrannical mind wants to push this false narrative that the computer is actually intelligent, when it is not.
And this is the danger -- a danger that Sam Altman (according to Elon Musk) does not care about whatsoever, which means he and others like him will continue to push for tyrannical control, using AI as the false tool to push a harmful agenda.
> Did we go to the Moon?
How would an AI computer answer this historical question?
We should never use AI to draw conclusions for us. It should not be used to make judgements nor to provide opinions.
AI should be used as a tool to gather as much data as it can provide but we need to trust our own human instincts and our own judgement. That is purely human domain.
If you want to know whether we went to the Moon or not you would ask the AI to provide evidence for both arguments and you can further drill into each of these evidences, until you have a good idea to be able to make that judgement yourself.
If you really want to answer this question, i would say that it would take anywhere from a week to a few weeks to do a thorough investigation using Grok.
Here is a good starting point. As you can see there are strong evidence for both sides of the argument, which means you have a lot of work cut out to come to the conclusion of either one or another. You will have to spend days drilling into each of these evidences and further figuring out what it takes to refute each of them on both sides and see what stands for sure.
Once you have eliminated all the impossibilities, whatever option remains, no matter how improbably, is the truth. And AI will definitely help you do that as long as you have the time, energy and patience for this.
What are some of the strongest evidence that points to the fact that the Apollo indeed land on moon?
Thought for 39s
1. Lunar Retroreflectors Apollo astronauts from missions 11, 14, and 15 placed laser ranging retroreflector arrays (LRRRs) on the Moon's surface. These devices reflect laser beams fired from Earth back to their origin, allowing precise measurements of the Earth-Moon distance. Independent observatories worldwide, such as the Observatoire de la Côte d'Azur in France, McDonald Observatory in Texas, Apache Point Observatory in New Mexico, and Haleakalā Observatory in Hawaii, have successfully detected and used these reflectors since their deployment—including detections made shortly after the missions (e.g., Apollo 11's reflector was detected by Lick Observatory on August 1, 1969, while the crew was still returning). The reflectors continue to function today, providing ongoing data for scientific research. This evidence is irrefutable because it requires physical hardware on the Moon, and the precise photon return patterns match small, man-made objects rather than natural lunar surface reflections.
2. Moon Rock Samples The Apollo missions returned approximately 382 kilograms (842 pounds) of lunar rocks and soil, which have been analyzed by hundreds of independent scientists and laboratories around the world (unaffiliated with NASA). These samples exhibit unique characteristics impossible to replicate on Earth or fake, such as an age of up to 4.5 billion years (older than any Earth rocks by about 200 million years), a composition lacking water or atmospheric exposure, and isotopic ratios matching the Moon's origin from solar system materials. For instance, a rock from Apollo 17 was dated to 4.417 billion years old by researchers at Curtin University in Australia. The samples also closely match those collected by the Soviet Union's unmanned Luna probes, providing cross-verification from a rival space program during the Cold War.
3. Images from Independent Lunar Probes Non-U.S. spacecraft have photographed the Apollo landing sites, showing clear evidence of human activity such as disturbed soil, descent stages, rovers, and astronaut footprints. Examples include:Japan's SELENE (Kaguya) probe in 2008, which captured images of lighter, disturbed dust at the Apollo 15 site, matching Apollo-era photos.
India's Chandrayaan-1 (2008) and Chandrayaan-2 (2019) orbiters, with the latter releasing a high-resolution image of the Apollo 11 Lunar Module descent stage at Tranquility Base in 2021.
China's Chang'e 2 probe (2010), which identified traces of Apollo hardware and rover tracks.
South Korea's Danuri orbiter (launched 2022), which has imaged Apollo sites.
These international missions, conducted by countries with no incentive to perpetuate a U.S. "hoax," provide visual confirmation independent of NASA.
4. Independent Tracking and Monitoring During Missions The Soviet Union, a geopolitical rival, closely tracked all Apollo missions using radar and intelligence equipment from their Space Transmissions Corps, confirming NASA's reports without dispute—if faked, they would have exposed it for propaganda value. Amateur radio operators and observatories worldwide also independently detected transmissions and calculated orbits. For example:Bochum Observatory in Germany recorded Apollo 11 signals without the typical NASA communication artifacts.
Jodrell Bank Observatory in the UK tracked Apollo 11 alongside the Soviet Luna 15 probe.
Radio amateurs like Larry Baysinger in the U.S. and others in Europe captured live astronaut transmissions using home-built equipment.
Additionally, the Soviet RATAN-600 radio telescope detected signals from Apollo Lunar Surface Experiments Packages (ALSEPs) in 1977, matching NASA's reported coordinates and power outputs.
5. Photographs from NASA's Lunar Reconnaissance Orbiter (LRO) While NASA-operated, the LRO (launched in 2009) has provided ultra-high-resolution images of all Apollo landing sites, visible to the public and verifiable by experts. These show descent modules, scientific instruments, rover tracks, and footprints persisting in the airless lunar environment. Combined with the international probe images above, this forms a comprehensive visual record that debunks claims of staging, as the sites match historical Apollo data exactly.
What are some of the reasonable doubts to consider that there may have been attempt to fake the moon landings?
Thought for 33s
1. The American Flag Appears to "Wave" in a Vacuum Skeptics point to video footage where the flag planted on the Moon seems to flutter or move as if in a breeze, which they argue is impossible in the Moon's airless environment. This suggests the scene was filmed on Earth in a studio with wind or fans. The doubt arises from the intuitive expectation that fabric should hang limp without air, leading some to question if the entire extravehicular activity (EVA) was staged.
2. No Stars Visible in Photographs Apollo photos show a black sky with no stars, despite the Moon lacking an atmosphere that scatters light. Theorists claim this is evidence of a hoax, as stars should be brightly visible, and their absence might indicate NASA forgot to add them to a faked backdrop or that the images were taken under studio lights on Earth. This doubt seems reasonable to some because everyday experience on Earth (e.g., stargazing) contrasts with the photos.
3. Inconsistent Shadows and Lighting Shadows in Moon photos often appear at varying angles or lengths, which skeptics interpret as signs of multiple artificial light sources (like spotlights in a film set) rather than a single source (the Sun). Some also note "hot spots" or uneven illumination, suggesting close-range lighting rather than distant sunlight. This raises doubts about the authenticity of the lunar environment, as it mimics common movie production errors.
4. Survival Through the Van Allen Radiation Belts The Van Allen belts are regions of intense radiation around Earth, and theorists argue that 1960s spacecraft lacked sufficient shielding to protect astronauts, implying the missions were faked to avoid lethal exposure. This doubt is bolstered by claims that film cameras would have been fogged by radiation, ruining photos. It seems plausible given Cold War-era nuclear tests (like Starfish Prime in 1962) that intensified the belts, raising questions about NASA's risk assessments.
5. No Blast Crater Under the Lunar Module The Lunar Module's descent engine should have created a visible crater or scattered dust from the rocket thrust, but photos show the surface largely undisturbed beneath it. Skeptics see this as evidence of a prop lowered by wires in a studio, not a real powered landing. This doubt persists because it aligns with expectations from Earth-based rocket tests, where exhaust often leaves marks.
6. Crosshair Anomalies in Photos Cameras used reticles (crosshairs) etched on glass for scale, but in some images, these appear behind objects (e.g., rocks or astronauts), suggesting the photos were composited or altered. Theorists argue this indicates "pasting" elements onto backgrounds. It raises reasonable suspicion for those unfamiliar with photographic artifacts, as it implies post-production tampering.
7. Identical Backgrounds in Separate Photos Photos purportedly taken miles apart show nearly identical mountain backdrops, leading skeptics to claim a limited painted set or backdrop was reused. This doubt questions the scale and authenticity of the lunar landscape, as distant features should vary more noticeably.
Both of these prophecies were partially true, partially false. A lot of jobs will become redundant. Especially the low end coders and you absolutely want these jobs to become redundant. But at the same time, it will expand what we as humans can achieve and create newer, higher level jobs that focus more on human thinking, intelligence, creativity.
That’s a little misleading, don’t you think?
Yeah, it’s true new jobs will be created as others are lost, but on the Ai timeline, “at some point”, most go away.
create newer, higher level jobs that focus more on human thinking, intelligence, creativity.
How are you defining jobs in this statement? Will they exist within the traditional framework of what we currently see understand a job to be?
*a paid position of regular employment.
*a task or piece of work, especially one that is paid.
????
Personally, I don’t see 95-99% of humans collecting a check “from an employer” 20 yrs from now. When we take what you said to its logical conclusion
“the beauty of AI is that the number of different problems it can be used to solve is limitless.”
“There will always be better and better, even open source”
We end up with Ai doing just about everything and what it doesn’t do/do better will probably be that way just bc we decided it should be that way and not because it can’t do it better🤷♂️
All said,,,,, humans will work in the future, but it will be work based on the raising of the world’s consciousness. Not sure yet what that entails, but my gut is telling me it’s so..
I specifically gave you two similar examples - robotics and Internet and drew my observations from what we have actually seen. So no, not misleading because its supported by human experience.
but on the Ai timeline, “at some point”, most go away.
I have no idea what this means.
How are you defining jobs in this statement?
In 80s and 90s I would have had a hard time defining what new jobs would be created while the Internet took away tons of jobs. Because these jobs did not exist yet. Many concepts and many technologies hadnt even developed yet, even though they would develop as Internet became prevalent.
Same is true with AI. We cannot even imagine the kind of jobs that would exist in 20-30 years let alone 50-100 years.
Will they exist within the traditional framework of what we currently see understand a job to be?
I can only hazard a guess, based on history again that the kind of jobs would transcend whatever frameworks we have right now. Whatever the new jobs would be, it would allow humans to live a purposeful and independant life - whether they earn money from an employer, whether they earn money for their own ventures or whether they do something entirely different we cannot imagine right now.
Personally, I don’t see 95-99% of humans collecting a check “from an employer” 20 yrs from now.
I agree. And this is a good thing and part of human liberation. We went from 80% Americans self employed in 1800s to "working for wages" as the most common means of earning income in 1900s. The tide is going to turn back and we will all be liberated from wage slavery.
We end up with Ai doing just about everything and what it doesn’t do/do better will probably be that way just bc we decided it should be that way and not because it can’t do it better
Very valid concern. This is why I keep harping that "regulation of AI" should not put AI under monopoly of elites. It has to be accessible to everyone for ever. Even if not perfectly, atleast like what happened with Internet.
Even with all the censorship on facebook, twitter, reddit etc, we still had sites like GAW that helped us circumvent these censorship because of Internet.
Similarly, as long as everyone is able to wield the power of AI and train their own models, no one can monopolize what problems can and cannot be solved. Each is limited only by their own imagination.
All said,,,,, humans will work in the future, but it will be work based on the raising of the world’s consciousness. Not sure yet what that entails, but my gut is telling me it’s so..
but on the Ai timeline, “at some point”, most go away.
Meaning,, looking down down on a 20-50 year timeline starting from when ChatGPT was first introduced to the public you would see when humans were almost completely removed from the workforce.
It possibly could happen sooner but the amount of time needed to manufacture enough humanoids to cover those jobs and then integrate them seems like a two decade project🤷♂️. Even if it could be done I don’t think it would be a healthy scenario for humanity.
But,,,, all bets off if we’re told of an in coming cataclysmic event!
Garbage data. Reddit, Twitter, Facebook, ETC.... trained on the best data we have online... Just pull up a sewage truck and pump that into your data center to train on.
Very true. Most normal people don't realize that one of the heaviest uses of AI is for what's politely called "erotica," which is just straight-up extreme pornography of the worst kind. Includes both stories and pictures. The AI gets trained on that, too, 24/7.
*Writers groups are filled with posts from "writers" who want to know how to get the various AI programs to write/draw/paint this stuff for them. And they're succeeding in forcing it to do so.
It sounds like you have a strong opinion. Curiously is it founded in actual know, or just gut instinct ? What’s your age and background in AI? I work in tech infra and older. I don’t think AI is possible to be sentient without quantum but I think you’re far off.
When I first started going online in the mid 1990's there was no ONE search engine. You had a multitude of choices and some were better than others.
Also, the internet back than was vastly different than what we have today. It was a interconnected web of sites. People made sites about things that interested them. People also linked to other sites they found interesting. Now, the internet is Google and a hand full of apps (Twitter, Facebook, TikTok).
I hate all posts that say “look here what I got grok to say” All totally retarded as you can get these AIs to come to ANY conclusion you want with your prompts. Stop posting that crap, frens.
Tell the mod that stickies his own posts fellating AI. DarQ2Light i think. Seems like every day he puts something that he "asked grok" or "chatgpt's opinion."
Honestly wish I wasnt the only one calling out his bs. I can't "deport" either because he's a mod.
I think AI has some powerful possibilities, and, like any tool, can be wielded to do amazing things or do amazing evil.
I also think AI is way over-hyped. It can't, or at least shouldn't, be used everywhere. Right now, people in leadership positions who don't fully understand AI are pushing their underlings to shoehorn it into their organizations, resulting in a bubble of hiring and spending that is eventually going to burst.
I think it's going to end up like the Internet but on an accelerated scale. Low acceptance at first, then a huge adoption full of promising and diverse possibilities, followed by corporate control being seized and it devolving into a crappy version of it's former self that we're all stuck with and can't fix.
Agreed. I also think sometime between 1 and 5 years from now we will start seeing companies losing tons of data, IP, and competitive edges because of stupid ideas like letting Microsoft Copilot onto their LAN and allowing it to index all their data "to help their employees and businesses be more efficient". Companies are allowing Copilot to have access to many - if not all - docs, even while they are being created. Even EMail. Microsoft was bad enough with Windows, but Copilot is an awful idea for businesses.
Your concerns are understandable, especially given the sensitivity of enterprise data.
However, with proper governance, Copilot is designed to operate securely without compromising data, IP, or competitive edges.
When correctly configured, Copilot accesses only the Microsoft 365 data (docs, emails, etc.) a user is explicitly authorized to view via the Microsoft Graph, respecting existing permissions.
Data stays within the organization’s tenant, encrypted at rest and in transit, and isn’t used to train Microsoft’s models or shared externally. Prompts and responses are processed in-memory, and admins can audit interactions via Microsoft Purview without Microsoft accessing the data.
With robust controls such as strict access policies, regular audits, limited data scope, and third-party plugin vetting, Copilot doesn’t index or expose sensitive data beyond what’s permitted.
For example, sensitive documents or emails remain inaccessible to unauthorized users, and web queries are anonymized, adhering to enterprise-grade protections like GDPR.
While misconfigurations could pose risks, proper governance mitigates these, allowing businesses to leverage Copilot’s efficiency benefits securely.
It’s critical to enforce these controls and train employees to use it correctly to avoid internal leaks or oversights.
It seems like AI is as reliable as Wikipedia. Garbage in, Garbage out.. I saw a photo of a cottonwood tree that accompanied an AI story.It was a large tree with an odd vine going up the trunk, high into the branches. The vine had white things on it, that I realized looked like cotton bols. AI is going to mis inform a lot of people. .
but perceptions of AI’s relentless march toward becoming more intelligent than humans, even becoming a threat to humanity, came to a screeching halt Aug. 7
No, thats the story pushed by the Cabal, because they know they have lost control of AI development.
As it happened, GPT-5 was a bust. It turned out to be less user-friendly and in many ways less capable than its predecessors in OpenAI’s arsenal.
Have jumped to GPT-5 exclusively since the day it launched. Use it exclusively for writing all my code, and its the best yet. Even companies like Cursor that use APIs for AI coding switched their defaults from Claude to GPT-5.
Saying that AI is going to be a dud is exactly like saying "Internet is going to be a dud" in the 90s.
I used to code using Grok exclusively and was extremely happy with it given it was free. Cost wise it was more efficient than Claude which was more capable but are through my credits like they were potato chips.
But after discovering OpenAI codex running gpt 3o, I completely stopped using Grok. The quality was similar I think, Grok kinda more humorous, but codex had the ability to with on large code base with lot of files plus it could match my coding style without breaking a sweat.
Now the new gpt-5 is absolutely stunning in its coding abilities. Surpasses Claude and no extra cost. But I need it to be put into codex so I can use it on large codebases. Right now it works on couple files a time
Maybe President Trump is damning AI with praise because he knows it is trash and that his critics will not be able to attack him when AI does fail.
Attacking Open AI for misspelling of states is a low blow because the states were smaller on the map used and this version of Open AI wasn't programmed to type smaller in limited space.
2 is definitely good observation. It has all the hallmarks of a "smear job" which always gives away what the enemy is really scared of.
As for 1 - when AI fails - I think AI has already succeeded beyond our wildest dreams. Does anyone here believe that this complex Q plan could have been worked out flawlessly without help of AI (perhaps far more advanced that whats publicly available) ?
I think Q is a military operation. It does have all the hallmarks of the military operation, especially when Donald Trump was recruited by the military to be President and when Q made his first appearance just over 9 months into POTUS's first term. I am saying that Q could not be an AI operation, but Q also talked about how the operation is up to three big movies. It could be likely that the mass arrests happened right under our noses and thousands of people are following a script or Q is simply a prophet looking to God for guidance.
I was listening to Defcon Zerq E005 this morning & Josh was asking Grok some questions about Q... and a very interesting thing occurred, indicating that it wasn't AI answering questions etc, but humans in this specific case ...
Very interesting... especially when you remember Amazon/Whole Foods "Just Walk Out" payment fiasco, allegedly powered by AI, but in reality it was 1,000 pajeets doing it instead via cameras etc.
Oh come on. Surely you subscribe to the theory of disruptive innovation? Let me pick your brain for ideas. That's how we utilize the strategy of the spoon and catch that low hanging fruit and be the game changers we know we can be. Our journey activated here and now with our thought leadership. We will sunset the stakeholders outside our paradigm by our brilliance.
Absolutely a Must Read. Throws much needed cold water on the overheated hype surrounding "Artificial Intelligence." Why you can't trust anything to a system that makes more mistakes than a bright third-grader.
I think the worst thing about AI is its name. Its a misnomer. Its got nothing to do with intelligence. If it were called, something like "Extremely powerful pattern learning / data crunching / analytical software" (not catchy, for sure) we wouldn't even be having these discussions.
AI should not be used to do any decisions that requires human insight or judgement. AI should be used as a tool with an extremely wide scope.
Added:
The correct term for these AIs is actually already there, I dunno how I missed it.
LLM = Large Language Model.
If we make it a point to call it LLM instead of AI, a lot of problems disappear.
Extremely good point. The triumph of marketing over realism.
Meanwhile, more conventional software is hooked up to mechanism and kills people without regard to consultation (737 MAX MCAS). I don't understand how people can feel comfortable about "self-driving" cars. Coming soon: air taxi service by robot drones. "Oh, the future is cool!" Yeah, it gives me a chill down my spine.
Long ago, I chanced across a science fiction story from the early years of pulp magazines, about automated automobiles. Instead of a mechanical brain, they had grown synthetic organic brains, suspended in a globe of nutrients, and hooked up to the vehicle controls. All goes well, until a (literally microbial) virus arises and begins to infect the synthetic brains with progressive insanity, with really bad results. The foreshadowing of the future a hundred years hence was uncanny.
I'm currently in a meeting by our development team praising AI and how much it can do for us. His woke ways keeps slipping out.. use chat GPT not Grok.. blah blah blah. He is also playing down how vulnerable the data being entered into the AI can be exposed. My boss already is shaking his head No about implementing it.
While handy, I can’t see paying for this. And that is the problem. Anything I were to use AI for will be public domain, so that means it can’t be used for work.
And most everything else, I can look up the data myself with better accuracy
The problem is deciding what is truth? You can use AI when it has been trained solely on the authoritative literature for your narrow subject area. In that case the truth is generally agreed upon. For generall use answering random questions it is just a fancier search engine that can be used to fool the gullible.
Nowadays, it feels like a conversation about artificial intelligence (AI) is always a stone’s throw away. Whether deeply entrenched in technology or casually reading the news, you will likely come across headlines and stories about the latest companies and developments in the AI space. These stories talk about all the great things that AI can do locally and globally, for individuals and businesses, as well as the world’s anxieties about adopting AI systems in their homes or workplaces.
https://coingeek.com/who-making-profit-in-artificial-intelligence-industry/
However, what is not talked about as often is who is making money in the AI industry.
Google’s (NASDAQ: GOOGL) latest earnings call brought this issue to the surface when the company revealed that its Q2 expenditure reached $13 billion, with most of that spending going toward data centers and services that support AI capabilities.
During the call, many analysts questioned whether the company’s multi-billion dollar investments in AI would turn a profit anytime soon. Google CEO Sundar Pichai said that AI is a long-term investment and that it is better to overinvest in AI now than to underinvest and fall behind. However, it is unclear when Google will begin to profit from its significant investment in AI—and they aren’t the only company in that boat.
OpenAI posted $540 million in losses the year it developed ChatGPT, attributing those losses to the spending required to train and develop GPT. Now, the company is allegedly on track to lose as much as $5 billion due to the costs associated with running the AI products it offers.
The high cost of AI: OpenAI’s financial breakdown
In OpenAI’s case, $4 billion has been earmarked for servers that maintain ChatGPT and the large-language models that power the chatbot. Another $3 billion is needed to cover the cost of training its AI models with new data, and $1.5 billion goes toward operations and the 1,500+ employees OpenAI has on staff.
Both small and large companies end up spending in these same areas—data centers, servers, and talent—when it comes to their AI operations. However, despite the large spending, these companies’ revenue is nowhere close to the costs they incur.
For example, OpenAI is estimated to generate around $3.4 billion in revenue, but that is nowhere near the ~$8.5 billion it spends to make its products possible.
AI industry’s biggest winners: How chipmakers and cloud providers benefit from AI spending
So where does all the money go that these tech giants are spending to get their AI products and infrastructure up and running? It goes to chipmakers, data centers, and cloud service providers. Every company in the AI industry needs at least one of these three items to run the AI arm of their business successfully. Very few companies can produce all of these elements in-house, and even if they can, the technology and bandwidth needed are ever-evolving, requiring the company to invest in the latest tech and infrastructure so they can grow, scale, and provide better products and services.
When it comes to chips, Nvidia (NASDAQ: NVDA) is the dominant player in that industry, with roughly 80% of the market share in the GPU space. AI companies need GPUs to train and run their models, and data centers need to accumulate GPUs so they can train and run AI models on behalf of their customers. This is why chipmakers have been the biggest winners in the AI industry so far.
When ChatGPT was released in November 2022, followed by a significant increase in competition in the AI industry, the price of Nvidia shares increased by over 500% in that two-year period. In the last year alone, Nvidia is estimated to have sold $34.5 billion in AI chips.
Simultaneously, data centers/cloud service providers such as AWS experienced significant growth in the sales of their cloud computing services, citing that their clients, especially those with businesses that had generative AI components, were a factor in Amazon’s (NASDAQ: AMZN) 17% year-over-year growth in its cloud computing service.
AI industry’s profit dilemma: High spending, low returns, and trouble on the horizon
It’s no surprise that infrastructure and ancillary services have been the biggest winners so far from a financial perspective. If a company is not fully vertically integrated, it must rely on third-party hardware, infrastructure, and service providers to deliver its goods or services. Relying on these providers isn’t unusual, but what’s giving investors in AI product providers concern is that it doesn’t look like the companies making these products will turn their big investments and expenditures into profit anytime soon.
Not only are most of the AI ventures of these companies currently not profitable, but the path to profitability remains unclear for many of them. These companies need to aggressively invest to stay competitive, pouring money into better hardware and server space so they can continue to train and run their AI models. Still, these products aren’t providing the revenue that exceeds the cost of bringing these products to life.
What is likely to make it even more difficult for many of these tech giants, who operate in silos and walled gardens, are the open-source offerings on the market, such as Meta’s (NASDAQ: META) Llama 3.1.
In an industry where many participants are not profitable yet, and the players in that industry are price-sensitive due to the high cost of operating and competing, a household name like Meta offering a high-quality free product (AI model) is likely to suppress prices further and lengthen the amount of time it takes for product providers to become profitable as players in the industry begin opting for Meta’s free, high-performing AI model to reduce their own costs and bring themselves closer to profitability.
All of this begs the question—is AI in a bubble?
The AI industry has experienced a meteoric rise in the last two years, and investment dollars have flowed into the space at a rate that does not appear to be sustainable. At the end of the day, investors want a return on their investment, and that return is nowhere in sight for most AI companies.
AI undoubtedly adds value to the world, improving business operations, leading to breakthroughs in math and science, and automating numerous workflows for individuals, freeing up time for them to focus on things that cannot be automated. However, only a few companies drive innovation in this highly saturated industry.
If these tech giants do not emerge with a clear revenue path in the AI industry, we are likely to see the industry’s rapid growth slow down and maybe even deflate to a level that reflects the real value these companies are delivering in the global economy.
In order for artificial intelligence (AI) to work right within the law and thrive in the face of growing challenges, it needs to integrate an enterprise blockchain system that ensures data input quality and ownership—allowing it to keep data safe while also guaranteeing the immutability of data. Check out CoinGeek’s coverage on this emerging tech to learn more why Enterprise blockchain will be the backbone of AI.
I’m more a movie/ headphones guy, but I like the way you think!
My largest position is in a fairly unknown biotech company “NWBO”, ever heard of it?
I have some GME and I’m a believer, but the price of movie and headphones has proven to be too tempting to pass up. Something tells me I’m going to be very happy I went that direction based on the amount I’ve been able to accumulate compared to where I would be had I been buying GME.
*for the longest there was a .01% chance in my head that the Q op could be a setup to screw us, but when I realized what GME was and then started noticing ALL THE OTHER PATHS being laid out,,, that thought flew out the window..
Things wrong with AI.
No scope.
No clear definition of problem/solution.
Garbage data. Reddit, Twitter, Facebook, ETC.... trained on the best data we have online... Just pull up a sewage truck and pump that into your data center to train on.
Economy. AI will take everyone's job! Yeah! Who will pay for AI?!?!?!
AI will take all the power!!!!!! You thought home prices were high! Wait till you see what we do with your power bill. AI will 100X your power bill!
Thank Elon and Sam Altman for total destruction of our society!
No consistency either. If a calculator gave you a different answer each time you asked the same Q, you would throw it into the trash. And not even merely divergent answers that at least feign a level of plausibility and correctness, but brainless hysterics.
When AI is gobbling up nonsense from other AIs into its LLM data sets, there is going to be crazy and escalating levels of unreality. The 1st gen ones had the benefit of consuming purely human datasets, like a pristine world that had never been mined before - easy and viable value extractions.
But what happens when the internet is 99% white-noise digital junk and very little actual data?
We already have dead internet theory. In 2015 50% of internet traffic was bots traffic and rising.
I dont see any reason why the quantity of email thats spam, proportion of phone calls that are scammers, proportions of twitter that are bots, proportion of the websites that search engines will never index as potential results etc wouldn't follow exactly how AI is going right now. The same data-trust problem repeated over and over in each new medium and niche. And forming a viable chain of trust over the top of the hostile and unrelenting flow of nonsense is the answer for each case.
And therein lies the problem -- and will ALWAYS be the problem.
WHO decides WHICH datasets are "correct?"
No computer can ever do that, which means it will ALWAYS come back to the FACT that a computer will be PROGRAMMED by a human -- who may or may not be honest or honorable, and will NEVER have enough knowledge to know which datasets should be accepted or rejected.
The AI will just spit out what some human told it to spit out, thus making "artifical intelligence" an oxymoron.
Human intelligence is enough of a problem already, and artificial intelligence is an impossibility.
Moreover, with scumbags like Sam Altman using their unjustly inflated ego to lie about what is really possible, we get a "tyranny of the 'expert'" problem -- just like the liars during the Covid scam.
Don't forget the countless people deliberately posting stuff to confuse any AI that scrapes it in order to sabotage it
Yes, because that is the fundamental mindset of someone like Sam Altman.
He is incapable of NOT thinking about how to control others -- just like any other standard issue libtard out there.
Their minds do not allow the concept of a free people making decisions that the libtard would not make, thus they think it must be "wrong" and needs to be "corrected" (i.e. re-educated).
Tyranny mindset. That's what these people are all about, and we should start calling it that to drive home the point.
Yes, exactly. Substitute "sociopath" for "libtard" and you've got it. Because that's what they are and all they care about is power and control.
Normal people care about love, beauty, and creativity.
Sociopaths care about power and control.
That's how you can tell the difference.
I think Altman has been associated with the philosophical/ social movement, “Effective Altruism”🤔
If you can consider what he’s heavily involved in, OpenAI, Oklo inc. (previous chairman), Tools for Humanity (co-founder), Worldcoin (co-founder), Apollo Projects, Y Combinator and Loopt AND you know what those things are about you can see why he’s being associated with Effective Altruism.
But,,,, he’s a major backer/investor in Reddit and depending on your views on why Reddit is the way it is you might surmise he’s the ultimate control wizard! 🤷♂️
I just know that Elon Musk was interviewed about a variety of topics, and the subject of AI came up. He said he talked about it at length with Altman, and asked about the down side of the harm that could be done.
Altman was oblivious and uninterested in talking about that.
Ergo, Altman cannot be trusted.
Yeah,,, in my mind that’s the psyop doing what a psyop does.
Dumbing everyone down
Ever considered humans have much more to give than just an 8 to 5 - - -?
Why have a spirit/soul whatever if you are destined to keep your head down for the majority of the day and work?
If you’re here you’ve broke free of a lot of what binds us as humans. But,,, it seems to me that the ultimate bind is how we have been sold the idea that we are meant too/better off if we work our ass off our whole life and then die.
Here’s the rub,,,, I LOVE working, but I don’t love working for others and having very little control over when my free time happens. I have no doubt that if I had more control over my time I would accomplish more “meaningful” things that would raise my consciousness and possibly that of others, ai seems to be the fix for that..
Anything you can reduce to strings of data and patterns can be used to train AI. Its not that it has no scope, but the scope is umlimited.
Just like the previous point - the beauty of AI is that the number of different problems it can be used to solve is limitless. Anyone trying to come up with a "definiton of problem/solution" is forcing themselves into a box.
AI is "garbage in, garbage out" technology. What differentiates different AIs is the quality of the training data. This is another beauty of this technology. There will always be better and better, even open source, AI models because anyone can train a model with their own data. Power to the people.
"Robots will take everyone's job"
"Internet will take everyone's job"
Both of these prophecies were partially true, partially false. A lot of jobs will become redundant. Especially the low end coders and you absolutely want these jobs to become redundant. But at the same time, it will expand what we as humans can achieve and create newer, higher level jobs that focus more on human thinking, intelligence, creativity.
Perfect forcing factor to unveil real energy solutions - be it safe nuclear or the so called Tesla energy.
It feels like you have never worked with software. The first step in creating a project that fails is to not clearly define the scope of work and to not limit requirements coming in.
We could give AI good clean data... where does that exist?
People need work and income. That is how our economy currently works. If AI can do everything why have any humans? Humans also need starter jobs to learn to work.
Power you are assuming the best case scenario. That the government will release the new (old researched in government lab) tech for this problem. I have a bridge to sell you... The bridge was in London at one point in time...
I think your perspective is coming from being a software contractor who takes requirements from a client and delivers working software.
Eventually you will get to see bigger picture of how not just a small piece of software is developed, but technology is created. Every single company that I worked in started off with a clearly defined problem, but somewhere along the way realised that their focus should be elsewhere and they pivoted. This is a standard lifecycle of not just startups but technology itself.
AI didn't pop out of nowhere. Precursor to AI was the field of Artificial Neural Networks (ANN), that started as early as 1950s. They had a clearly defined goal - to mimic the way neurons work in the brain and they had a well defined problem: Pattern recognition and classification.
When ANNs hit into the limitations of computational power, it was largely forgotten as a field and people moved on to more exciting areas. Deep learning algorithms - foundation of today's AI models - were just extremely complex versions of ANNs with much more optimized algorithms and the problem definition that can be solved by AI pivoted from pattern recognition to pattern generation.
All of today's AI models are nothing but pattern generators. You feed them billions of data patterns, and then you feed it a new pattern and it will extend it by generating the continuation of that pattern.
When you ask a question to Grok, all grok is doing is taking that string you typed and running it through its internal network and extending that string into something that, as a human, you perceive it to be an "answer" - but its just a string of letters.
So yeah, AI has a clearly defined function - pattern generation and extension - but what problems you can solve with this is unlimited.
To understand this concept that applies to all new technology, you can compare it with crypto space. When blockchains were originally created it was solving a well defined problem of decentralized consensus of financial transactions with no double-spending. Thats exactly what Bitcoin protocol implemented.
However, the problems you can solve with this is unlimited. Started off with currency, extended into programmable contracts, NFTs, and pretty much anything you could imagine - now reaching the problem of making liquidity available to people without banks as middlemen.
When we compared where we are now with where it started from, you cannot help but be amazed. And its the same with AI.
I guess you missed the point I was making, explaining how - just like with Robotics and Internet - while some jobs were lost, a lot more were made, and in the end we still have jobs for everyone - from newbies to veterans. However the jobs that newbies today do are different from the newbies did 50 years ago. And it will be the same with AI - because, and its important to keep in mind, AI is not a thinking entity. Its simply a pattern generator. We humans are the thinkers and we will always be that.
One thing I’ve learned from you guys is to never think in absolutes. That’s not to say there’s not absolutes that exist, but from my experience the moment I think something is set in stone is the moment it begins to change🤷♂️
*Do you believe the human mind creates reality? Literally,,, creates reality?? Are we only limited by our understandings??? If you believe that and accept it as truth then is it really that crazy to “”think”” that Ai will develop a level of creativity that matches/surpasses human ability?
*”If” our brains have the ability to receive/tap into the universal consciousness for answers/creativity (thinking in terms of frequency/radio waves) than maybe we can create something capable of this too!
I believe humans have a soul which connects if to the Universe or God. This is what differentiate is from even the most powerful AI. They can be smarter than is in many areas, sure but we'll never possess our instincts, insights, and spirituality that has guided us as humanity even through all the deception of Satan.
Who do you think Satan is?
What type of entity, if any at all??
It entirely depends on what your worldview about God is. If you believe God is a supernatural being as conventionally accepted in religious circles, then Satan is an angel who rebelled.
If you believe God is extra terrestrial beings who created our DNA and populated us here, then Satan is a ET who wants to treat us like cattle.
If you are an atheist - then God and Satan does not exist.
And so on.
False.
While in THEORY the scope might seem limitless, it will in reality ALWAYS have limits.
How would an AI computer answer this historical question?
Would it draw from an "approved database" of information?
Would it accept data from multiple databases that conflict with each other? How would it resolve the conflict? WOULD it resolve the conflict?
Humans cannot agree on how to resolve this and similar conflicts, so how would a computer that processes 1's and 0's do it?
If by "AI" we are ONLY talking about computers that can process data faster or more efficiently, then sure, it has its use. A computer programmer can write code much faster with AI -- and some companies are firing coders because they don't need them to do as much as before.
There will be some uses, for sure.
BUT ... human intelligence is all about (a) taking concrete ideas, then (b) imagining some abstract concept that can utilize those concrete ideas, then (c) applying that abstract concept back to existing and/or new concrete ideas.
This is what IQ tests measure: the ability to apply abstract concepts to concrete data.
Computers can ONLY process concrete data. They can also do it much faster and more efficiently than humans. BUT they CANNOT come up with NEW abstract ideas on their own. THAT requires human input.
Thus, they are not "intelligent," in a human sense; therefore, the term "artificial intelligence" is a LIE.
This is the crux of the problem, because the tyrannical mind wants to push this false narrative that the computer is actually intelligent, when it is not.
And this is the danger -- a danger that Sam Altman (according to Elon Musk) does not care about whatsoever, which means he and others like him will continue to push for tyrannical control, using AI as the false tool to push a harmful agenda.
Faster and more efficient to process data? Sure.
But that IS its limit.
We should never use AI to draw conclusions for us. It should not be used to make judgements nor to provide opinions.
AI should be used as a tool to gather as much data as it can provide but we need to trust our own human instincts and our own judgement. That is purely human domain.
If you want to know whether we went to the Moon or not you would ask the AI to provide evidence for both arguments and you can further drill into each of these evidences, until you have a good idea to be able to make that judgement yourself.
If you really want to answer this question, i would say that it would take anywhere from a week to a few weeks to do a thorough investigation using Grok.
Here is a good starting point. As you can see there are strong evidence for both sides of the argument, which means you have a lot of work cut out to come to the conclusion of either one or another. You will have to spend days drilling into each of these evidences and further figuring out what it takes to refute each of them on both sides and see what stands for sure.
Once you have eliminated all the impossibilities, whatever option remains, no matter how improbably, is the truth. And AI will definitely help you do that as long as you have the time, energy and patience for this.
That’s a little misleading, don’t you think?
Yeah, it’s true new jobs will be created as others are lost, but on the Ai timeline, “at some point”, most go away.
How are you defining jobs in this statement? Will they exist within the traditional framework of what we currently see understand a job to be?
*a paid position of regular employment.
*a task or piece of work, especially one that is paid.
????
Personally, I don’t see 95-99% of humans collecting a check “from an employer” 20 yrs from now. When we take what you said to its logical conclusion
“the beauty of AI is that the number of different problems it can be used to solve is limitless.”
“There will always be better and better, even open source”
We end up with Ai doing just about everything and what it doesn’t do/do better will probably be that way just bc we decided it should be that way and not because it can’t do it better🤷♂️
All said,,,,, humans will work in the future, but it will be work based on the raising of the world’s consciousness. Not sure yet what that entails, but my gut is telling me it’s so..
I specifically gave you two similar examples - robotics and Internet and drew my observations from what we have actually seen. So no, not misleading because its supported by human experience.
I have no idea what this means.
In 80s and 90s I would have had a hard time defining what new jobs would be created while the Internet took away tons of jobs. Because these jobs did not exist yet. Many concepts and many technologies hadnt even developed yet, even though they would develop as Internet became prevalent.
Same is true with AI. We cannot even imagine the kind of jobs that would exist in 20-30 years let alone 50-100 years.
I can only hazard a guess, based on history again that the kind of jobs would transcend whatever frameworks we have right now. Whatever the new jobs would be, it would allow humans to live a purposeful and independant life - whether they earn money from an employer, whether they earn money for their own ventures or whether they do something entirely different we cannot imagine right now.
I agree. And this is a good thing and part of human liberation. We went from 80% Americans self employed in 1800s to "working for wages" as the most common means of earning income in 1900s. The tide is going to turn back and we will all be liberated from wage slavery.
Very valid concern. This is why I keep harping that "regulation of AI" should not put AI under monopoly of elites. It has to be accessible to everyone for ever. Even if not perfectly, atleast like what happened with Internet.
Even with all the censorship on facebook, twitter, reddit etc, we still had sites like GAW that helped us circumvent these censorship because of Internet.
Similarly, as long as everyone is able to wield the power of AI and train their own models, no one can monopolize what problems can and cannot be solved. Each is limited only by their own imagination.
In this we both agree completely.
Mind Blown.
I was for sure that last one would be the moment this place went on the attack bc that was a pretty woowoo statement 😁
*Seems to me in the end it will have all been about raising consciousness
Meaning,, looking down down on a 20-50 year timeline starting from when ChatGPT was first introduced to the public you would see when humans were almost completely removed from the workforce.
It possibly could happen sooner but the amount of time needed to manufacture enough humanoids to cover those jobs and then integrate them seems like a two decade project🤷♂️. Even if it could be done I don’t think it would be a healthy scenario for humanity.
But,,,, all bets off if we’re told of an in coming cataclysmic event!
Very true. Most normal people don't realize that one of the heaviest uses of AI is for what's politely called "erotica," which is just straight-up extreme pornography of the worst kind. Includes both stories and pictures. The AI gets trained on that, too, 24/7.
*Writers groups are filled with posts from "writers" who want to know how to get the various AI programs to write/draw/paint this stuff for them. And they're succeeding in forcing it to do so.
There's a reason the term "Hentai Diffusion" circulates around AI communities
It sounds like you have a strong opinion. Curiously is it founded in actual know, or just gut instinct ? What’s your age and background in AI? I work in tech infra and older. I don’t think AI is possible to be sentient without quantum but I think you’re far off.
People acting like AI doesn’t suck because of all the censorship efforts. SMH. Gaslighting 101.
Same shit happened to Google btw, search engines are utter trash these days
When I first started going online in the mid 1990's there was no ONE search engine. You had a multitude of choices and some were better than others.
Also, the internet back than was vastly different than what we have today. It was a interconnected web of sites. People made sites about things that interested them. People also linked to other sites they found interesting. Now, the internet is Google and a hand full of apps (Twitter, Facebook, TikTok).
Agree. I remember dial-up BBS's and Finger/Gopher.
Lol, Ask Jeeves🐸
And Lycos.
I bet few people know that google is a number. It is 1 w/ 100 zeros after it.
And Archie and Reggie. I think there was even a Veronica...
I hate all posts that say “look here what I got grok to say” All totally retarded as you can get these AIs to come to ANY conclusion you want with your prompts. Stop posting that crap, frens.
Grok is just "Fact Check" w/ AI set up to defend the l;ies.
Tell the mod that stickies his own posts fellating AI. DarQ2Light i think. Seems like every day he puts something that he "asked grok" or "chatgpt's opinion."
Honestly wish I wasnt the only one calling out his bs. I can't "deport" either because he's a mod.
I'm of two minds on this.
I think AI has some powerful possibilities, and, like any tool, can be wielded to do amazing things or do amazing evil.
I also think AI is way over-hyped. It can't, or at least shouldn't, be used everywhere. Right now, people in leadership positions who don't fully understand AI are pushing their underlings to shoehorn it into their organizations, resulting in a bubble of hiring and spending that is eventually going to burst.
I think it's going to end up like the Internet but on an accelerated scale. Low acceptance at first, then a huge adoption full of promising and diverse possibilities, followed by corporate control being seized and it devolving into a crappy version of it's former self that we're all stuck with and can't fix.
Agreed. I also think sometime between 1 and 5 years from now we will start seeing companies losing tons of data, IP, and competitive edges because of stupid ideas like letting Microsoft Copilot onto their LAN and allowing it to index all their data "to help their employees and businesses be more efficient". Companies are allowing Copilot to have access to many - if not all - docs, even while they are being created. Even EMail. Microsoft was bad enough with Windows, but Copilot is an awful idea for businesses.
Your concerns are understandable, especially given the sensitivity of enterprise data.
However, with proper governance, Copilot is designed to operate securely without compromising data, IP, or competitive edges.
When correctly configured, Copilot accesses only the Microsoft 365 data (docs, emails, etc.) a user is explicitly authorized to view via the Microsoft Graph, respecting existing permissions.
Data stays within the organization’s tenant, encrypted at rest and in transit, and isn’t used to train Microsoft’s models or shared externally. Prompts and responses are processed in-memory, and admins can audit interactions via Microsoft Purview without Microsoft accessing the data.
With robust controls such as strict access policies, regular audits, limited data scope, and third-party plugin vetting, Copilot doesn’t index or expose sensitive data beyond what’s permitted.
For example, sensitive documents or emails remain inaccessible to unauthorized users, and web queries are anonymized, adhering to enterprise-grade protections like GDPR.
While misconfigurations could pose risks, proper governance mitigates these, allowing businesses to leverage Copilot’s efficiency benefits securely.
It’s critical to enforce these controls and train employees to use it correctly to avoid internal leaks or oversights.
It is currently only capable of searching the internet and then crafting a response that looks like straightforward english.
It seems like AI is as reliable as Wikipedia. Garbage in, Garbage out.. I saw a photo of a cottonwood tree that accompanied an AI story.It was a large tree with an odd vine going up the trunk, high into the branches. The vine had white things on it, that I realized looked like cotton bols. AI is going to mis inform a lot of people. .
Hahaha, LA Times sure wants that to happen.
No, thats the story pushed by the Cabal, because they know they have lost control of AI development.
Have jumped to GPT-5 exclusively since the day it launched. Use it exclusively for writing all my code, and its the best yet. Even companies like Cursor that use APIs for AI coding switched their defaults from Claude to GPT-5.
Saying that AI is going to be a dud is exactly like saying "Internet is going to be a dud" in the 90s.
Have you tried Grok for code? Saw some stats that it's way better at coding than GPT.
Although, it came from an X source, which may be biased.
I used to code using Grok exclusively and was extremely happy with it given it was free. Cost wise it was more efficient than Claude which was more capable but are through my credits like they were potato chips.
But after discovering OpenAI codex running gpt 3o, I completely stopped using Grok. The quality was similar I think, Grok kinda more humorous, but codex had the ability to with on large code base with lot of files plus it could match my coding style without breaking a sweat.
Now the new gpt-5 is absolutely stunning in its coding abilities. Surpasses Claude and no extra cost. But I need it to be put into codex so I can use it on large codebases. Right now it works on couple files a time
Two thoughts:
Maybe President Trump is damning AI with praise because he knows it is trash and that his critics will not be able to attack him when AI does fail.
Attacking Open AI for misspelling of states is a low blow because the states were smaller on the map used and this version of Open AI wasn't programmed to type smaller in limited space.
2 is definitely good observation. It has all the hallmarks of a "smear job" which always gives away what the enemy is really scared of.
As for 1 - when AI fails - I think AI has already succeeded beyond our wildest dreams. Does anyone here believe that this complex Q plan could have been worked out flawlessly without help of AI (perhaps far more advanced that whats publicly available) ?
You're on to something,
For all we know, the text in most Q posts might have been written using AI assistance; Perhaps even future prediction technology was utilized!
This is just speculation though
I think Q is a military operation. It does have all the hallmarks of the military operation, especially when Donald Trump was recruited by the military to be President and when Q made his first appearance just over 9 months into POTUS's first term. I am saying that Q could not be an AI operation, but Q also talked about how the operation is up to three big movies. It could be likely that the mass arrests happened right under our noses and thousands of people are following a script or Q is simply a prophet looking to God for guidance.
Q being a military operation, and Q using AI for precision planning of this operation are not mutually exclusive.
There is also behavioral physically as the military can predict what happens in the future.
West Wigina sounds rather feminine to me🤣😂
A few days ago I said the AI bubble was going to pop, and now this article.
3 6 9 and 17??
I was listening to Defcon Zerq E005 this morning & Josh was asking Grok some questions about Q... and a very interesting thing occurred, indicating that it wasn't AI answering questions etc, but humans in this specific case ...
Very interesting... especially when you remember Amazon/Whole Foods "Just Walk Out" payment fiasco, allegedly powered by AI, but in reality it was 1,000 pajeets doing it instead via cameras etc.
Like each technology bubble, some ICs swooned over buzz words and pumped money into something that didn't need IC.
And here we go again.
I absolutely hate buzzwords. They can take their buzzwords and put them all back into that box they want you to think outside of.
Oh come on. Surely you subscribe to the theory of disruptive innovation? Let me pick your brain for ideas. That's how we utilize the strategy of the spoon and catch that low hanging fruit and be the game changers we know we can be. Our journey activated here and now with our thought leadership. We will sunset the stakeholders outside our paradigm by our brilliance.
Now, does anyone know what we do here?
Absolutely a Must Read. Throws much needed cold water on the overheated hype surrounding "Artificial Intelligence." Why you can't trust anything to a system that makes more mistakes than a bright third-grader.
I think the worst thing about AI is its name. Its a misnomer. Its got nothing to do with intelligence. If it were called, something like "Extremely powerful pattern learning / data crunching / analytical software" (not catchy, for sure) we wouldn't even be having these discussions.
AI should not be used to do any decisions that requires human insight or judgement. AI should be used as a tool with an extremely wide scope.
Added:
The correct term for these AIs is actually already there, I dunno how I missed it. LLM = Large Language Model.
If we make it a point to call it LLM instead of AI, a lot of problems disappear.
Extremely good point. The triumph of marketing over realism.
Meanwhile, more conventional software is hooked up to mechanism and kills people without regard to consultation (737 MAX MCAS). I don't understand how people can feel comfortable about "self-driving" cars. Coming soon: air taxi service by robot drones. "Oh, the future is cool!" Yeah, it gives me a chill down my spine.
Long ago, I chanced across a science fiction story from the early years of pulp magazines, about automated automobiles. Instead of a mechanical brain, they had grown synthetic organic brains, suspended in a globe of nutrients, and hooked up to the vehicle controls. All goes well, until a (literally microbial) virus arises and begins to infect the synthetic brains with progressive insanity, with really bad results. The foreshadowing of the future a hundred years hence was uncanny.
I'm currently in a meeting by our development team praising AI and how much it can do for us. His woke ways keeps slipping out.. use chat GPT not Grok.. blah blah blah. He is also playing down how vulnerable the data being entered into the AI can be exposed. My boss already is shaking his head No about implementing it.
If they really want to implement some LLM, insist on a locally managed LLM.
While handy, I can’t see paying for this. And that is the problem. Anything I were to use AI for will be public domain, so that means it can’t be used for work.
And most everything else, I can look up the data myself with better accuracy
The problem is deciding what is truth? You can use AI when it has been trained solely on the authoritative literature for your narrow subject area. In that case the truth is generally agreed upon. For generall use answering random questions it is just a fancier search engine that can be used to fool the gullible.
We need more Human Intelligence (HI), not a flood of questionable AI.
I pray AI dies and doesn't come back.
Nowadays, it feels like a conversation about artificial intelligence (AI) is always a stone’s throw away. Whether deeply entrenched in technology or casually reading the news, you will likely come across headlines and stories about the latest companies and developments in the AI space. These stories talk about all the great things that AI can do locally and globally, for individuals and businesses, as well as the world’s anxieties about adopting AI systems in their homes or workplaces. https://coingeek.com/who-making-profit-in-artificial-intelligence-industry/ However, what is not talked about as often is who is making money in the AI industry.
Google’s (NASDAQ: GOOGL) latest earnings call brought this issue to the surface when the company revealed that its Q2 expenditure reached $13 billion, with most of that spending going toward data centers and services that support AI capabilities.
During the call, many analysts questioned whether the company’s multi-billion dollar investments in AI would turn a profit anytime soon. Google CEO Sundar Pichai said that AI is a long-term investment and that it is better to overinvest in AI now than to underinvest and fall behind. However, it is unclear when Google will begin to profit from its significant investment in AI—and they aren’t the only company in that boat.
OpenAI posted $540 million in losses the year it developed ChatGPT, attributing those losses to the spending required to train and develop GPT. Now, the company is allegedly on track to lose as much as $5 billion due to the costs associated with running the AI products it offers.
The high cost of AI: OpenAI’s financial breakdown
In OpenAI’s case, $4 billion has been earmarked for servers that maintain ChatGPT and the large-language models that power the chatbot. Another $3 billion is needed to cover the cost of training its AI models with new data, and $1.5 billion goes toward operations and the 1,500+ employees OpenAI has on staff.
Both small and large companies end up spending in these same areas—data centers, servers, and talent—when it comes to their AI operations. However, despite the large spending, these companies’ revenue is nowhere close to the costs they incur.
For example, OpenAI is estimated to generate around $3.4 billion in revenue, but that is nowhere near the ~$8.5 billion it spends to make its products possible.
AI industry’s biggest winners: How chipmakers and cloud providers benefit from AI spending
So where does all the money go that these tech giants are spending to get their AI products and infrastructure up and running? It goes to chipmakers, data centers, and cloud service providers. Every company in the AI industry needs at least one of these three items to run the AI arm of their business successfully. Very few companies can produce all of these elements in-house, and even if they can, the technology and bandwidth needed are ever-evolving, requiring the company to invest in the latest tech and infrastructure so they can grow, scale, and provide better products and services.
When it comes to chips, Nvidia (NASDAQ: NVDA) is the dominant player in that industry, with roughly 80% of the market share in the GPU space. AI companies need GPUs to train and run their models, and data centers need to accumulate GPUs so they can train and run AI models on behalf of their customers. This is why chipmakers have been the biggest winners in the AI industry so far.
When ChatGPT was released in November 2022, followed by a significant increase in competition in the AI industry, the price of Nvidia shares increased by over 500% in that two-year period. In the last year alone, Nvidia is estimated to have sold $34.5 billion in AI chips.
Simultaneously, data centers/cloud service providers such as AWS experienced significant growth in the sales of their cloud computing services, citing that their clients, especially those with businesses that had generative AI components, were a factor in Amazon’s (NASDAQ: AMZN) 17% year-over-year growth in its cloud computing service.
AI industry’s profit dilemma: High spending, low returns, and trouble on the horizon
It’s no surprise that infrastructure and ancillary services have been the biggest winners so far from a financial perspective. If a company is not fully vertically integrated, it must rely on third-party hardware, infrastructure, and service providers to deliver its goods or services. Relying on these providers isn’t unusual, but what’s giving investors in AI product providers concern is that it doesn’t look like the companies making these products will turn their big investments and expenditures into profit anytime soon.
Not only are most of the AI ventures of these companies currently not profitable, but the path to profitability remains unclear for many of them. These companies need to aggressively invest to stay competitive, pouring money into better hardware and server space so they can continue to train and run their AI models. Still, these products aren’t providing the revenue that exceeds the cost of bringing these products to life.
What is likely to make it even more difficult for many of these tech giants, who operate in silos and walled gardens, are the open-source offerings on the market, such as Meta’s (NASDAQ: META) Llama 3.1.
In an industry where many participants are not profitable yet, and the players in that industry are price-sensitive due to the high cost of operating and competing, a household name like Meta offering a high-quality free product (AI model) is likely to suppress prices further and lengthen the amount of time it takes for product providers to become profitable as players in the industry begin opting for Meta’s free, high-performing AI model to reduce their own costs and bring themselves closer to profitability.
All of this begs the question—is AI in a bubble?
The AI industry has experienced a meteoric rise in the last two years, and investment dollars have flowed into the space at a rate that does not appear to be sustainable. At the end of the day, investors want a return on their investment, and that return is nowhere in sight for most AI companies.
AI undoubtedly adds value to the world, improving business operations, leading to breakthroughs in math and science, and automating numerous workflows for individuals, freeing up time for them to focus on things that cannot be automated. However, only a few companies drive innovation in this highly saturated industry.
If these tech giants do not emerge with a clear revenue path in the AI industry, we are likely to see the industry’s rapid growth slow down and maybe even deflate to a level that reflects the real value these companies are delivering in the global economy.
In order for artificial intelligence (AI) to work right within the law and thrive in the face of growing challenges, it needs to integrate an enterprise blockchain system that ensures data input quality and ownership—allowing it to keep data safe while also guaranteeing the immutability of data. Check out CoinGeek’s coverage on this emerging tech to learn more why Enterprise blockchain will be the backbone of AI.
Soooo you’re saying I should sell my Ai/quantum stocks, sit that money on the sidelines and wait for the crash??? 🤔🫡
Follow what Trump is doing. He says we're going to be the AI capital of the world. That's a good clue to hodl.
My other recommendation is to look into GME.
I’m more a movie/ headphones guy, but I like the way you think!
My largest position is in a fairly unknown biotech company “NWBO”, ever heard of it?
I have some GME and I’m a believer, but the price of movie and headphones has proven to be too tempting to pass up. Something tells me I’m going to be very happy I went that direction based on the amount I’ve been able to accumulate compared to where I would be had I been buying GME.
*for the longest there was a .01% chance in my head that the Q op could be a setup to screw us, but when I realized what GME was and then started noticing ALL THE OTHER PATHS being laid out,,, that thought flew out the window..