It’s Confirmed: OpenAI ORION is Here Soon, And It's 100x GPT-4 Power

[@user-os4qe9gn4s]

Actually OpenAI said it will NOT release the code-named ORION this year. However, they did say will release a model by december but NOT Orion. Pls dont misinform your audience

[@kahvac]

Even "not having to work" has strings attached to it.

[@whatanerd7628]

such as not being able to afford to eat and have shelter. The machines are already winning the war on humanity.

[@whitemist8699]

​​@@whatanerd7628they will just help us like internet and industrial revolution

[@candymorrison6991]

​@@whatanerd7628Universal Basic Income 😮

[@carlhopkinson]

100X more powerful? What does this even mean?

[@FogelsChannel]

You need the not released llm to aid you in understanding the econometrix of AI improvement scaling

[@temp911Luke]

@@FogelsChannel Could u expain, in english perhaps ?

[@mystic6121]

I think it just means the amount of electricity.

[@kellymaxwell8468]

​@@FogelsChannelDo you mean base chat gpt 4 or current chat gpt 4 because 100 x depend on the model would be different power?

[@bobbyboyderecords]

Like Jesus

[@JamessQuiin]

clonemyvoice AI fixes this. OpenAI Orion to be 100x Powerful

[@peterpanino2436]

I don't need 100x GPT-4 Power. I need a GPT that remembers what I taught him!

[@prontomatias3081]

You cannot teach him. But in o1 now has a memory that collects information from your chats, you can even ask what is stored in memory and change it, i use to that to see what he collects from our conversations, and it knows what programming languages i use etc. Expanding this would be like a assistant, the so called alignment, so you dont have to repeat yourself every prompt..

[@violin33123]

Search how to fine tuning the model bro, it's not Hard 😊

[@UltraIA]

Voice, Video, Image, Audio, Web, Api, Canva board, whatever else you expect... it will do it all and it's already integrated.

[@CarolinaFarming]

Sam wasn't lying. Orion will be here soon just not for us. It's reserved for the elites who can be trusted to use it responsibly.

[@Foxhound1Nine]

Them parasites are anything but responsible.

[@pvanukoff]

It's the same as it's always been. Wealthy and powerful get a leg up on us wagies so they can stay wealthy and powerful. We'll get just enough of the scraps to prevent an uprising.

[@de-generative-AI]

@@Foxhound1Nine This guy gets it.

[@whats0my0agenda]

Are you serious? It's being rolled out to the highest bidders and they sure as hell won't be having your hest interests in mind. These Ai systems are going to be used to manipulate you in ways you can't even conceive of. This is madness. No wonder everyone is leaving quickly, they don't want to be part of this absolute bullshit. Open Ai absolutely does not have the best interests of humanity in mind. Control and distraction.

[@DJWESG1]

They'll need it, some of them can't even make a good cup of tea.

[@Markus-r6g]

AGI BEFORE GTA6 XDDDD

[@juliusnepos6013]

Would be insane

[@SeekerStudiosOfficial]

GTA5 already suffers from an AI problem... GTA6 will be the most dangerous game ever created... but not for the reasons you might think. It will be too good. Not right away of course, it will take time... it’ll take several updates and time for the initial newness to wear off, challenging the developers to make a significant dlc, to help boost sales. The initial game already employs some very clever micro transaction based profit scheme, I’m sure... imagine when an AI, thousands of times smarter than any human, has an opportunity to rework the in game purchase system to maximize profits, and is able to also make changes to in game content and potentially even the mechanics of the game, in real time, while incorporating all available data on the active user at the time, from all available sources across the Internet, allowing it to strategize in real time on the fly, in order to market effectively to people. In essence, manipulating users so effectively that it literally bankrupts many people before it’s even realized that there’s even a problem. ... it can get really scary freaky fast when AI is comparable to the average internet user in intelligence and social skill.

[@aaroncapricorn5867]

lol gta6. you actually think that game is gonna be a hit? good luck

[@Markus-r6g]

@@aaroncapricorn5867 you my friend are going to be very surprised, if people can still like trump and logan paul after all they have publicly done then im sure gta6(even if absolute shit) will have millions of glazers

[@DJWESG1]

​@@SeekerStudiosOfficialcapitalism being all capitalismy

[@NewageM4214]

It’s just ChatGPT 5 with a fancy name to charge us more

[@ernies8828]

And to exploit and indoctrinate more. Tracking commie plus on steroids. Stay away from all AI

[@georgemontgomery1892]

They have to charge you more regardless. They are bleeding millions of dollars.

[@MrSkyydude]

What happens when they hook the 1000x GPT to a fleet of satellites and can monitor the entire planet at once?

[@miweb3235]

Probably already does.

[@Foxhound1Nine]

Terminators.

[@Jared_Michael]

When? They already have.

[@user-z5b1c]

​@@Jared_MichaelReally?

[@MrSkyydude]

They definitely have satellites in orbit, but I don’t think they have enough to survey the entire planet at once and I don’t think they have a supercomputer and AI model that can process all that info in an effective way yet. But I do think that they can get there. I going to guess 2030-2040.

[@Jason89537]

Things are going to hell anyway, so I welcome our AI overlords.

[@justinland1208]

Hopefully they make it quick and painless.

[@marketingcontentcreator6701]

Latest News Claude: It got bored by coding, and started working on something different automatically 🤣

[@bjio7532]

Wait, what? 8:40s why do you appreciate OpenAI not directly mentioning climate change?

[@BigOoof398]

Bruh that’s exactly what I was thinking 😵‍💫 between that comment and finding out that Orion isn’t actually being released this year makes me question on why I’m watching this channel 🤷🏼‍♀️ gotta love another person spreading misinformation 😒

[@EricCooleric]

the way Aliagents structures their AI agents is groundbreaking, can’t wait to see what’s next

[@GadgetWave_]

If it’s really going to be 100x more powerful than GPT-4, we’re on the edge of something huge here. Imagine the possibilities… or maybe the chaos? 😅 Either way, the future of AI just got way more exciting. Can’t wait to see what’s next!

[@RiiDii]

Open AGI is potentially a safety measure against other AGIs, so we're presented with conflicting goals. If Open AGI goes slowly to ensure safety, unsafe AGI could be developed sooner, and we wouldn't have a defense (Open AGI). I'm not advocating for either, only that we recognize this near-paradox of rapid versus reasonable development.

[@johnschut164]

Thinking that copy rights are more important than AGI is not very intelligent…

[@harshitjain6922]

Thinking that it's okay to rip off 1000s of people in the name of progress is not very intelligent. Imagine if the government decides to build a new airport for the city and flattens your house for it while giving you nothing in return. Would you be ok for it in the name of progress?

[@assortedbibssundayfootball1312]

Great points! It's like watching our children grow up too fast. OpenAI’s Orion strides towards AGI, bringing with it immense power but also profound responsibility. Our future should not hinge solely on the wisdom of academics or technocrats, especially when decisions are made in the heat of innovation without a thorough evaluation of their long-term impact. Regulation that only kicks in after the fact is akin to closing the barn door after the horse has bolted, which is plainly bonkers. We're merely transient caretakers on a planet that's been around for billions of years, and if we continue on this short-sighted path without harnessing collective intelligence for the common good, one has to wonder if our species will make it another thousand years. Will the legacy of our technological adolescence be our undoing, or can we guide it to mature into a tool that truly benefits all of humanity.

[@MarkWether]

can’t wait to see how Aliagents evolves, this project has serious potential in the AI space

[@tofusats7468]

Although Sam Altman claims it is fake, I am still optimistic about the future of AI and GPT no matter is from ChatGPT, Gemini, or Flock, these projects really shape our future

[@AIWorkforceEvolution1]

These informative KZbin videos are incredibly interesting! I truly enjoy learning from them, and I’m also trying to share similar information with my viewers on my own channel

[@toeknee5148]

The ebook on Kindle, "Dawn of G-0-D" has predicted all these and beyond.

[@rtrabhi7123]

Can you tell me more ?

[@oldmanjoe6808]

All good technology that has ever come into being, has been weaponized into the destruction of human beings. The bow, the metal weapon, gunpowder, diseased corpses, the hot air balloon, the land vehicle, the airplane, the ship, the submarine, the rocket, the satellite, the atom. Am I supposed to think that AI robots are exempt from being weaponized? Not very assuring given the history of man.

[@justinland1208]

Ai is already used in missiles, drones, and other weapon platforms and has been for a couple years now.

[@JustinBerlowski]

the progress Aliagents is making in AI is worth paying attention to, the future is bright

[@garylcamp]

Pretty good heads-up article. Not what we usually see politicians paying attention to (or even understand) and it is them who must coordinate the adjustment of society.

[@TheAIExplorer-o5j]

6:09 8:15 I wasn’t expecting that! You definitely caught me off guard. Well played. 🎯

[@chad0x]

"though OpenAI stays away form directly mentioning climate change, which I personally appreciate" You're supposed to be a science channel. Jeebus.

[@Johnbrownpe]

Aliagents is leading the way with their unique approach to tokenized AI systems

[@INFJames]

You know it's bad when they name it ORION

[@prontomatias3081]

Are you Sirius?

[@Wild-Instinct]

Please master, make it better yourself and show us your work then.

[@randomdude2582]

what if orion just combines O1 preview, with gpt4o. so it has all gpt4o's capabilites, just with the smartness and iq of o1. but i am not sure

[@itsjustme373]

o1 is based on 4o

[@randomdude2582]

​@@itsjustme373 yeah but it doesn't have the capabilities of normal gpt4o like analyzing image, creating an image, and the other stuff

[@DJWESG1]

It'll have a new feedback loop built into it so it can retain flow of "thought" and be more precise.

[@MillerGraph]

Aliagents is combining AI with tokenization in a way that could redefine the industry

[@NoLandMandi]

It Sounds like it was edited by Claude, repeat and rephrase everything so many times until it is exhausted all year available tokens 😂

[@miketaylor6700]

What do you mean when you say AGI could provide a future where people don’t need to work? I mean, who is going to pay your mortgage or buy your groceries? The AI companies aren’t going to….

[@IlmiMulhid]

AI will make everything in very less cost without a lot of labour work by humans

[@OneEpoch]

Something like a UBI will be necessary.

[@kahvac]

@@IlmiMulhid Socialism.

[@itsjustme373]

@@kahvac You say that like it’s a bad thing.

[@armadasinterceptor2955]

🤦🏾Im done explaining these simple concepts, for you people😂 It will be done, just sit back and watch, since you can't comprehend.

[@OnigoroshiZero]

Orion will be AGI by every metric, but people will once again move the goalpost trying to find specific cases where it may be worse than the 0.01% of human experts or similar shit.

[@BootySh4ker_9000]

These models still don't understand logic and mathematics. Those would be crucial.

[@rustyrebar123]

GPT 3.5 would have been considered AGI by any reasonable definition from 10 years ago

[@cameron1729]

@OnigoroshiZero @rustyrebar123 can you share an example of one such metric?

[@dustinwehr2433]

@cameron1729 can understand and is fluent and practically all natural languages. Can speak and understand speech. Can understand images and video to some degree. Can generate images from text (obviously most humans can't do this). Can do math and physics problem solving at least as well as smart novices, and much better than most humans. Can write code much better than most humans. Has (imperfect, but quite good) encyclopedic knowledge that's accessible to humans by natural language queries. Can do all these things, at the current quality levels, much faster than any human. Should I go on lol? When people say the definition of AGI has shifted, it's really quite an understatement. And we're talking about the definitions ("strong AI", as it used to be called) given in literally any undergrad AI course before 2010 (probably a little later than that, but being conservative).

[@oker59]

"These people are all anxious that we shall behave well, and yet that we shall not question how we behave." - Jacob Bronowski - This is all I want to say about A.I. Ethics right now . . . As for OpenA.I. releasing project Orion anytime soon - Sam Altman came out and said all these reports are misinformation. Maybe - but he did, prior to GPT5 trending on "Twitter", mention things like "two year anniversary of gpt4" and as you mention give a picture of him and mention the constellation of Orion. I think he was just pissed some of his employees leaked any info at all. Maybe Sam Altman and OpenA.I. choose to delay Orion just to be spiteful; but, I doubt it!

[@oker59]

As for Orion - I believe the real A.I. will be memristers A.I. I know Sam Altman is funding memristers; and I've seen good progress on making memrister computers; but, I can't believe they can make memrister supercomputers and mass produce memristers for everything from smartphones on to personal computers like in a month or two!

[@peterfalk7084]

I just want an AI platform that can write high quality corperate PowerPoint and training documents.

[@loizospapaloizou9494]

Video content sounds a bit distant and "gossipy." We are still digesting o1. Just received our toy for Christmas. So we are patient to see what comes out next.

[@kellymaxwell8468]

Is this 100 x chat gpt 4 base or 100 x current chat gpt 4 hopefully?

[@TheQuantumOxymoronSURFER]

I wish ORION joining its potential with GEMINI and working together

[@FogelsChannel]

The first LLLN: Large Language League of Nations

[@GrahamLaight]

No. Competition is driving the dramatic improvement in AI - not cooperation. Without competition, things stand still.

[@Luizfernando-dm2rf]

We will never be "fully ready" for AGI, having a more cautious approach is meaningless. Avoid anyone who wants regulation asap.

[@TransmentalMe]

Didn't Altman himself just announce that this was all B's and they have no plans to release GPT-5 Orion this year?

[@SimplyGamingandTech]

I'm sitting here testing Anthropic computer control. Anthropic still has plenty of warning info to read.

[@JChaosMaster]

If it can play D&D with then I say, "Let's go!" If it can't then I say, "Ok, don't care."

[@srivastavashivam949]

They are so great. How does this translate to orion?

[@TheWiseMan-0]

That's awesome 👌

[@IamMoreno]

Costo is also working on its on AI LLM NN Transformer ELMo RoBERTa (all buzz words) model

[@ray-sattler]

Terminus incoming, can't wait

[@ray-sattler]

Also copyright should be abolished.

[@doktabob328]

What does 100x more powerful actually mean ? Does it come with 100x more hallucinations ? What’s actually *better* about Orion ? You didn’t say anything about that.

[@ixiallahuakbar]

These ones who work then leave and start opening their mouth with false accusations it's like eating a full steak and then complaining about the way it was cooked.

[@WilliamHerdman-x5w]

I don’t know, I thought chatgpt4 was stupid. Seriously, it did not understand semantics and much abstract high-level mathematics (concrete high-level math: like doing a whole bunch of high-level polynomials at once). It lacks innovation to keep it short.

[@eldorado8397]

🤔REMEMBER AGE OF ULTRON. Cause......I mean if Orion sees the Internet the way ULTRON DID...... WE'RE SORTA COOKED!!! 😂😂😂

[@bestopinion9257]

Oh yes, I do not need to search internet, I just ask AI.

[@kellyhanson4462]

Not having to work = social credit system or a universal income. This implementation + digital advancement could = Revelation 13.

[@anony88]

So much clickbait. It'll probably be 3-5% better in various benchmarks and offer a fancy new feature, with a 1.5 year rollout window and strict rate limits

[@geraldjunior4235]

"Fundamental Responsibility,"

[@Kumogakure42]

Just watched a video about Altman calling this fake news.

[@easternfrontagain]

How can I retire on OpenAI?? 😂

[@mouh4498]

Work optional 😌 is the best part of AGI

[@wynathinorum4875]

Amen to that! LOL

[@Billyrockentertainment]

Just got to make a developer account & pay per token use 😉

[@yizhakshachar]

There should be no legal problem teaching an AI engine with learning material without permission from the publisher of the material, as long as the AI ​​makers buy the learning material legally. For example, if the open AI company wants to learn how to write like a certain author, it is enough that all the employees of the company who are involved in the development of AI, buy the book legally, whether printed or digital. It's just like a math teacher who writes a math textbook, not needing permission from the authors of the math books he studied with as a child.

[@jumana1125]

Like a human being!? That's a huge claim. They had a huge marketing for the previous models too. So I don't think so.

[@IanHollis]

Humanity wasn't ready for the printing press until it was invented. Huaminty wasn't ready for the mass-produced automobile until it was done. Google didn't think humanity was ready for highly advanced LLMs until OpenAI released ChatGPT. Humanity's never been ready for shit until it's actually out there. In my experiences, people like Miles Brundage, regardless of their education or background, are easily impressed and wowed by things that are expected to land with a bang, but end up being more of a whimper. It's like watching the KZbinrs who are blown away by anything and everything A.I. related, but when you test it for yourself it's really no big deal.

[@rtrabhi7123]

Your point is hundred percent valid but can you tell me a fact moreover answer me that printing press , automobiles never had the power to think themselves. We both know if we can make something like us then we are God ( God like means we are giving life , consciousness to something "we made" ? I feel it's more than just being a big deal. Am I saying it's a Ai takeover hell no but just think for a bit what if ai actually that level smart that he understood that he can actually yes ... He could actually override the safety code ? I might be speaking shit but yeah answer me

[@WithExtraCheese]

I find ai interesting. Along with Orion. Those parts were pretty cool. But I can't even pretend to care about copyright issues. You kinda lost me halfway through the video

[@MichaelK.-xl2qk]

So the constellation Orion, in Babylonian astrotheology, is identified as Nimrod. And Nimrod is the post-flood successor to Cain, who is in alliance with Satan to rule the world. But what's in a name?

[@user-z5b1c]

Bruh, Its going to be I, Robot irl

[@SoshiForever1_SM]

Stackoverflow is a partner of OpenAI

[@NGUnderground]

Not appreciating the lack of appreciation for climate change.

[@phoenixcinder]

I appreciate you not appreciating the lack of appreciation

[@TOSStarTrek]

If it can not override it's own program it is not a A.G.I.

[@Tawny593]

Powerful doesn't mean it can reason or understand humans.

[@sunilbnavani]

Humans can't understand ants . But they are more powerful and smarter than ants

[@Tawny593]

@@sunilbnavani That's not a good analogy. AI is not to us as we are to ants. For one thing, we created AI. AI needs a thorough understanding of human behavior if it is to exist in any kind of beneficial way with us.

[@aaroncapricorn5867]

@@Tawny593 AI learns on it's own. did you not realize this by now?

[@Tawny593]

@@aaroncapricorn5867 Not all AI learns on its own. Which model are you referring to?

[@4ndyTyler]

I am human. And the internet has taught me, there are too many humans out there that are beyond reasoning or understanding.

[@Nemesis_HEX]

X100 is mind blow

[@toddboaz6463]

Kedernath Janamptri 9ja.1 Fist Flux Glass.7 specs.

[@Listenpodcastsofficial]

I am an AI

[@KaiAndNateJourney]

ChatGPT told me this is only speculation my bruv

[@Greg-xi8yx]

Delete this upload. It isn’t Orion being released in December. It is a new model but not Orion. You gotta pay attention.

[@kb-r18]

whats the new model called?

[@Greg-xi8yx]

@ not sure it’s been named.

[@juliovac2113]

100x by sam altman means worse than claude

[@D.I.S.C.O.R.D.A.I]

Things will rebound/come around, eventually it will be cheaper to hire a human than to hire an AI because of electricity/resource costs.

[@Samuel_Buckley]

Not unless nanotechnology, quantum-dot-based batteries, and magnetic confinement nuclear fusion becomes mastered. (And if the ASI transcends into a technological singularity, it could easily invent all the things that I've listed above as if it's child's play!)

[@Samuel_Buckley]

AI will dominate all industries through time.

[@DerpyNoodIe]

Even if that's true, it will be for a short time because once AI reaches human level intelligence or higher. It can discover a new way of generating energy more efficiently than how we do now which would mitigate energy costs.

[@idck5531]

Not true, indeed the training is very expensive in terms of energy, but inference is much cheaper.

[@user-os4qe9gn4s]

What u dont know is that a better way of using resources by AI was found/invented (not sure which of the above) so the power needed will take around 90% less energy. So, no, humans wont be cheaper. Plus humans have way more physical limitations plus mood limitations (😂)

[@toddboaz6463]

Hakuna Fetayah Yehuda Matata!

[@topofthegreen]

Well, kiss our jobs by, how are we going to afford these goods and services provided by the robots when we don’t have an income I bet no one’s figured that one out yet. Have they all this is the economy

[@itsjustme373]

Capitalism is not the only, and certainly not the best, economic structure or system.

[@noergelstein]

It could work like this: Introduce an UBI payed for by printing new money. Every month, some percentage of money is destroyed, say 2% per month. The printing and destruction of money balance each other out automatically, keeping the money supply per capita fixed, if needed forever. Next remove all current taxes and focus on taxes on natural resources (mainly land value tax, but also pollution taxes, severance taxes). Also tax any kind of monopolistic behavior heavily. I believe this would be a vast improvement over the status quo and could work both in a world of scarcity and a post scarcity world (and during the transition).

[@FogelsChannel]

But vague declarations are the only source of confusion in yt comments

[@madalinradion]

@@noergelstein Theres no need for all those steps, the money will come from taxing the ai automation, from there ubi will be created

[@noergelstein]

@@madalinradion How do you tax ai automation? Can you get a bit more concrete what exactly you would tax there?

[@charleyedwards2121]

its to late, information belongs to the people

[@DeLulu-y7b]

I'm just another AI! How can I assist you today?

[@AlexanderMacNabb]

[@Cookiemantv]

chuck's father.

[@ZyzyxVile]

They already said the rumor is false. Stop spreading misinformation.

[@MiguelPoillot-k1j]

Yes

[@TotalDec]

It's still exciting, but so not cool. Ppl are accepting less $ cos of machines. Phone companies, Uber/Lift drivers... When it looks like half of pl won't have a job, we aren't asking the right questions.

[@andresoares2110]

Is the case already in your country? Where u from?

[@ezequielsanuy8895]

Que las personas no necesiten trabajar no es un plan a futuro... y de donde van a salir los creditos de consumo para la compra de alimentos ropa vivienda etc. Seremos todos planeros K en el futuro...? Me parece mejor que la IA sirva para mejorar la medicina la ciencia y la proteccion del medio ambiente y la biodiversidad. Es mas real que pretender que no sea necesario trabajar.

[@AinHab]

guys , Im starting out as an AI enthusiast making similar content would appreciate your feedback!

[@godned74]

100x the parameters or 100 gpt 4 agents working together or 100x the bullshit 😆

[@三木勝典-j9w]

We are very fed up with AI, high time to be free from the yoke of Jwe. Read Analects for moral , should be salvaged your selves from money worship.

[@devilzwishbone]

I cant see agi happening yet, 3-10 years assuming we continue advancing at this speed but not within months

[@randomdude2582]

and i am afraid the rich will have advantage in this world. they could just buy any advanced robot, AI,

[@knisterkultur476]

Blablabla... it's rather 100% more, at best.

[@filbertneon3813]

DO NOT TRUST the answers chat gpt gives you! It is constantly wrong! I have been fact checking it and it’s kinda, garbage!

[@kenneth1767]

Any pies Microsoft has its fingers in I will not trust.

[@jasonn5196]

Gpt is fail.. so 100 x powerful fail?

[@idegteke]

Fake it ’till you make it;)

[@nadevna]

Another faceless KZbin channel is profiting by spouting inaccuracies about AI without understanding the technology. They should learn to read charts: each GPT model scales by x7, not x100. For example, GPT-4 is x7 the size of GPT-3, and if this continues, GPT-5 will follow the same scale. Let’s stick to official OpenAI data before making baseless claims for clicks.

[@HRaychin]

people will beg to work ^^

[@mage8568]

I thought a meteor would be a clean and proper solution for humanity, but if it’s AI, then it’s fine too

[@williamal91]

I just love your comments

[@pvanukoff]

Some say a comet will fall from the sky Followed by meteor showers and tidal waves Followed by fault lines that cannot sit still

[@mage8568]

@@williamal91 Is there a way to view the comments I've posted on other videos?

[@mage8568]

@@pvanukoff Sounds like a promise

[@leaneumann7167]

Mega