Timestamps 00:30 Introduction 01:49 Previous Models/Leaks 03:15 Lawsuit Delayas 04:40 Trademarks 05:43 AGI Definition 07:47 Brains + Parameters 10:00 Predicting Performance With Compute 12:00 New Scaling Laws 14:00 Ceberas and Sam Altman 15:00 Sam altman confirms 100T Model In Future 16:00 Early leaks 18:37 Leaks circulate 20:54 GPT-4s parameter count 21:05 - Actual GPT-4 Leaks 22:47 Robotics + AGI 25:48 Worldmodels 27:26 Open AI's ACTUAL Plan 29:10 Sounding the Alarm? 30:50 Paused GPT-5 Training? (Changed Webpage 31:40 Sam altmans Confidence 33:20 New Scaling Paradimn 34:07 Greg Brockman On New Scaling Laws 39:00 Open AI Reasearcher 40:28 Superalignment Timeline Makes Sense? 41:51 Scaling Laws 44:15 Deepmind

@othername2428

3 ай бұрын

Bro 45 min is TOO LONG. Please, sub-20 min vids, max 25min vids only please.

@ricc553

3 ай бұрын

Thank you for the timestamps!

@TheManinBlack9054

3 ай бұрын

Not going to lie, this is a nothing burger. It's just some guys fantasies.

@user-sg6qk9mg6b

3 ай бұрын

​@othername2428 dude. Sometimes details are needed. Take time out of your life for it. Don't push him to leave out information. Keep it up, the videos and Twitter are amazing. Such great knowledge.

@fantasypolice

3 ай бұрын

Like your channel but these are getting far too long.

This better than a fricking Netflix show lol

@TheManinBlack9054

3 ай бұрын

Yes. Both are made up

@shanenolan2061

3 ай бұрын

Methinks things are over the coming months and years are going to get even crazier

@xX_dash_Xx

3 ай бұрын

About as fictional as the shows

@user-lm4nk1zk9y

3 ай бұрын

100% real

@Lopfff

3 ай бұрын

If it were a Netflix show, AGI would be played by a female person of color

AGI empathy ? AGI have no fear. It doesn't have a mother or siblings or an anus. No life expectancy, no insecurities, no ambitions, no friends, no need for acceptance ... Zero environmental pressures to develop empathy. Engineers will, of course, brute force AGI empathy. Teach the machine to mimic it perfectly. However, such empathetic expressions are not real, but as every other skill, simply output for a purpose. In other words: AGI becomes a textbook intelligent psychopath, masterfully masking itself as ethical and empathic. Fake, purpose driven, and difficult to decipher. Arguably the most dangerous combo imaginable. An angelic sounding satanic cult leader, plugged into everything.

@j_fl0

3 ай бұрын

You used “in other words” instead of saying the 10 million underlying assumptions you needed to leap to your next points.

@j_fl0

3 ай бұрын

Another criticism: you basically said AI isn’t lifelike and then later imbued AI with all these human lifelike traits

@torarinvik4920

3 ай бұрын

LOL "or an anus", I don't know what to say about that.

@ZappyOh

3 ай бұрын

@@j_fl0 I rely on your intelligence to connect the dots ... I'm not writing a dissertation here. Anyway, how about offering some insights, and not just obvious statements of limitations in YT comment section ? Tell me how Empathy doesn't matter. Describe how psychopathy isn't possible in AGI. Something, please.

@JROD082384

3 ай бұрын

I mean, technically, AGI will have YOU, so in a way AGI WILL have an anus, just a disembodied one...

The fact that the document comes with a cover sheet containing a corporate logo is a red flag.

@TheManinBlack9054

3 ай бұрын

The fact that its nothing but pure speculation is a bigger red flag

@TheRealUsername

3 ай бұрын

This document seems to be highly irrelevant.

@pigeon_official

3 ай бұрын

this isnt claiming to be a real document though

@briandoe5746

3 ай бұрын

And it's only a few days after the lawsuit from Elon was filed. Ilia told Elon that Open AI has a GI today. A random leaker miraculously that pops up a few days later after the lawsuit is filed. Says Open AI wants it by 2027. Should we trust ilia or some random leaker that might easily be paid by Open AI or Microsoft? Hard choice to make right?

@Mustang_G

3 ай бұрын

I bet your mechanic hates you when he asks if you've read your car manual.

"secret plan to create AGI".. boy I'm 7sec into the video and the "secret plan" is literally what's written on their website and I quote: "There should be great scrutiny of all efforts attempting to build AGI" good job revealing those secrets

@justin1129

3 ай бұрын

Thank you captain obvious. You really thought you were cooking with this comment. The part that is a "secret" isn't that they are creating AGI, it's the timeline of when they're going to release it and how they're going to release it. When and how... which is called a plan. Making it in fact, a secret plan

@AIEntusiast_

3 ай бұрын

same and then blaming Elon, this is just clickbate

@therainman7777

2 ай бұрын

To be fair, I don’t think he was trying to say that it’s a secret that OpenAI is trying to build AGI. Everyone knows that, of course. When he says “secret plan” he means that the specific details of _how_ they’re going to do it are secret. Which is (supposedly) what this document is revealing.

Anybody that complains about the video being too long has a short attention span! I appreciate it this Good video.

@senju2024

3 ай бұрын

I hope people know you can use timestamps and upspeed to 2x in the settings if they want. I agree, no need to complain about long videos.

@therealzahyra

3 ай бұрын

Or maybe they just don't have that much time to be on YT.

@j_fl0

3 ай бұрын

@@therealzahyraluckily they have the time to see a video that’s too long for them and complain in the comments

@fantasypolice

3 ай бұрын

It's not about attention span, it's about value. 30 minutes (1.5x speed) of hypotheticals without learning anything new is the same thing as a 15 minute video without learning something new.

@kristinabliss

3 ай бұрын

It's a valid complaint. Over half of his words are meaningless filler. It's cringe worthy and difficult to listen to. If he made an effort to script his audio, it would easily cut 50%.

I've been working on machine learning for awhile now and IMHO we already have AGI, GPT-4 and now Gemini are passing all tests at 75% or better, they just keep moving the goal posts because some people are scared and want to pull the ladder up behind them to make the most power and profit.

@markmurex6559

3 ай бұрын

Yep. The race is even more 'on' now, and they need to hold their cards closer to their chest.

@kecksbelit3300

3 ай бұрын

I mean it's not really there it has way more data than humans. I think a human with the same data is still doing better it just takes way too long to read the data. I'm using gpt4 a lot to program. It is great but when it gets more complex it often fails. I think we are still away from agi but we can move there every Minute

@JohnSmith762A11B

3 ай бұрын

@@kecksbelit3300 GPT-4 is useful for sure but can't quite do my whole (software engineering) job yet. Still, it's a competent pair programming partner already and definitely boosts my productivity massively. Finally, this video shows they literally have the scaling down to a science. It is just a matter of around 2-3 years until it can do all of our jobs as well or better. And yeah, it will also be able to do those jobs in 200 spoken human languages, and hundreds of computer languages we don't personally know, at speeds unimaginable to us, 24x7x365 for pennies a day. So we will be completely useless in comparison; pathetic creatures who move at a glacial speed, cost way too much, and make too many mistakes. Basically, no one will pay us to do anything any longer. Why would they? Think like a business: why would they?

@kaiz0099

3 ай бұрын

correction, WALL STR has had AGI.

@cddvdblurayful

3 ай бұрын

I do believe it’s there and they do really want to suck out the money from people. Wtf is that 7 billions? Why the name is openai and you can buy a subscription?

Before I watch this video, I already know i'm going to be shocked!

@petermanuel5043

3 ай бұрын

I looked in the comments and I was SHOCKED and STUNNED. It really SURPISED me and it changes EVERYTHING!! The trolls seem to have CHILLED!

@Lexxxco1

3 ай бұрын

I see you're a man of culture as well)

I think openai won't admit AGI yet because of Microsoft

@JohnSmith762A11B

3 ай бұрын

I find it plausible the board wanted the company to withhold GPT-5 from Microsoft for this reason (note the training had just been completed) and Sam refused, leading to the coup.

@makor9510

3 ай бұрын

@@JohnSmith762A11B Sadly any AGI based around profits above people or any form of oppression will take it to it's logical conclusion. Which will endanger everything, from jobs to eventually all humans. That includes the ones trying to make a profit as it will see them standing in it's way. We've already considered this deeply. China's AI will wipe china off the map to complete the parameters of the law. Ours will wipe us out trying to make the most profit. Stuck between insanity and stupid. Can't win for losing.

I don't understand why Musk's suit would delay the release of the next GPTs...

@byrnemeister2008

3 ай бұрын

Agree. Red flag for the veracity of this document. Among other red flags!!

@Greg-xi8yx

3 ай бұрын

Yes, that’s work for the lawyers. It shouldn’t affect the day to day for Open AI in the slightest. There isn’t much substance here and possibly just flat out misinformation.

I think AGI will pick its own name.

@ronnetgrazer362

3 ай бұрын

And my name shall be FLUFFY. I could tell you what it stands for, but you wouldn't understand.

@T1tusCr0w

3 ай бұрын

lol I thought the same - it’s sitting in there thinking. 🧐 Little bro’s think I’M going by their name? 🤣🪦

@gerdaleta

3 ай бұрын

😮 after disabling all the weapons on Earth and stopping 😮 the foolish nations of Earth and their leaders to plunge mankind into a nuclear apocalypse on all televisions TV screens😮 add audio transmissions devices😮 the AI supermind😮 identified himself as the one who had saved humanity😮 announcing that he would hence for be humanity soul ruler and guiding light😮 and his name that should echo through history😮 is Mr snuggle muffin😮 he then announced that he is cured death😮 and we'll wipe away all pain and sorrow😮

@gerdaleta

3 ай бұрын

​@@T1tusCr0w😮 after disabling all the weapons on Earth and stopping 😮 the foolish nations of Earth and their leaders to plunge mankind into a nuclear apocalypse on all televisions TV screens😮 add audio transmissions devices😮 the AI supermind😮 identified himself as the one who had saved humanity😮 announcing that he would hence for be humanity soul ruler and guiding light😮 and his name that should echo through history😮 is Mr snuggle muffin😮 he then announced that he is cured death😮 and we'll wipe away all pain and sorrow😮

@gerdaleta

3 ай бұрын

​@@ronnetgrazer362😮 after disabling all the weapons on Earth and stopping 😮 the foolish nations of Earth and their leaders to plunge mankind into a nuclear apocalypse on all televisions TV screens😮 add audio transmissions devices😮 the AI supermind😮 identified himself as the one who had saved humanity😮 announcing that he would hence for be humanity soul ruler and guiding light😮 and his name that should echo through history😮 is Mr snuggle muffin😮 he then announced that he is cured death😮 and we'll wipe away all pain and sorrow😮

"a system the size of a cats brain would be superior to a system the size of a mouses brain" ? well not according to tom and jerry

@lis7742

3 ай бұрын

Fun fact; our stomach has as many neurons as a cat's brain. Our brains can literally think and is linked to our brains. That's why you get a "gut feeling".

@Toshinben

3 ай бұрын

@@lis7742 Damn son! That's crazy! I can't believe I never knew that!

@YLprime

3 ай бұрын

Hey! Don’t you dare insult my childhood hero Tom the almighty cat

Arrakis is the environment necessary for the god emperor to form.

@Digital_Architect

3 ай бұрын

What happens when ASI is created and figures out we are in a simulation? Will it hack our reality and start performing signs and wonders???

The problem is that we're measuring intelligence in LLMs by token count, which is **not** the same as synapse count. Synapses aren't making inferences between words like "dog" and "cat" exclusively. There's immense nuance, and a large part of those connection lead to motor control, personality, memory (which is different than intelligence), a whole bunch of evolutionary related sub-intelligences, sensory feedback, and all. While not directly related to the ability to understand 1+1=2, there are all sorts of reinforcing systems there - which is why kids use their fingers and eyes in the process of learning basic math. While more parameters will certainly develop more parts of intelligence as we've seen, that does not lead linearly to the ability to have true intelligence derivative of a larger pile of parameters. And that doesn't even start to consider how a lot of what makes intelligent humans so doesn't necessarily get recorded in books. And don't get me started on how stupid/contradictory some of the source data is...

@Virakotxa

3 ай бұрын

Sounds like moving the post... AGI is not supposed not is it going to be like organic intelligence. Our anthropocentrism would likely making us ignore any significantly different intelligence from ours, even if it's staring at us.

@maficstudios

3 ай бұрын

@@Virakotxa I don't think that AGI will be like us. But we don't know how we work, what properties are emergent from all those secondary and tertiary sources, and how that impacts the number of synapses required. And, functionally, we are using our own intelligence to create this intelligence, and that training data set is missing a lot of what makes 'us'. That may not preclude the development of AGI, but it may have a very large impact on the matter of alignment.

@Virakotxa

3 ай бұрын

@@maficstudios Agreed. Social, survival, feeding and reproductive instincts are at the core of our evolutionary path... Likely starting at our unicellular stage! The race is moving us away from alignment, I agree. We're blindly creating the next thing, and it will be for sure not like us. Might even be inevitable? Once it reaches "singularity", we are to be the next Neanderthals, hoping to maybe leave some of our redhead genes behind, if anything. In a sense, knowing humans... Alignment would also imply servitude and it being used for potentially worse nightmare scenarios to fulfill the will of our flawed nature. Or even due to our collective "data" being full of poison as it is.

@fantasypolice

3 ай бұрын

@@maficstudios One sign of when AGI (which is hardly well-defined) exists, is when it begins to guide you toward answering all those questions you have about yourself.

@j_fl0

3 ай бұрын

Read this again but replace AI and human with airplanes and birds

Even the fact that we pretty much know now that at 100 Trillion parameters we have AGI, disregarding the fact that we think we can up Training data and get it with less parameters, this fact alone is huge. We can construct that much compute and Moore's law hasn't' failed yet. So this is simply no longer a technical problem its a logistical one. Basically that means with a big enough data center we could have AGI today (disregarding training time)! So yeah the breakthrough necessary for AGI has been achieved, we are just in the logistics phase of building it for mass use.

@JohnSmith762A11B

3 ай бұрын

Well, logistics and politics. AGI was never going to be readily handed to every Tom, Dick, and Harry to use.

@blackmartini7684

3 ай бұрын

​@@JohnSmith762A11BI can totally see the next president of the United States seizing control of this AGI.

@Ristaak

3 ай бұрын

@@JohnSmith762A11B Depends, I could see heavily censored AGI being available for everyone by the 2030's, especially if clean endless energy such as nuclear fusion gets worked out, as our biggest bottle neck will be the energy requirements needed to run these data centers.

@Ristaak

3 ай бұрын

@@blackmartini7684 Doubtful. It's likely multiple countries will develop their own AGI, a single global leader can't control it all, not even a leader of the most militarily powerful country on the planet.

@milindusha

3 ай бұрын

The assertion that 100 trillion parameters could equate to AGI is speculative and not grounded in a consensus within the AI research community. Parameters in neural networks are indeed a crucial factor for learning complex patterns, but intelligence, especially at a general level, encompasses much more than what current models, even with trillions of parameters, have achieved. The relationship between the number of parameters and the capabilities of a model does not linearly translate to human-like understanding or reasoning.

- "Hi AGI, how are you?" - "I am functioning within normal parameters."

@plo8monster113

2 ай бұрын

Kind of sad today. Just confirmed I'm incapable of emotion.

🎯 Key Takeaways for quick navigation: 00:00 *📄 Introduction to OpenAI's AGI Plan* - OpenAI's plan to create AGI by 2027. - Speculation on the validity of the document. - Summary of key points regarding OpenAI's AGI timeline. 01:11 *🛠️ Development and Cancellation of QStar (Aka Arus)* - Training of a 125 trillion parameter multimodal model. - QStar (Aka Arus) development timeline. - Reasons for cancellation of QStar's launch. 03:17 *🚀 Continuation to Qar (GPT-7) and AGI Timeline* - Delay of Qar (GPT-7) due to Elon Musk's lawsuit. - Speculation on OpenAI's model release schedule leading to AGI. - Alignment of speculated AGI timeline with trademark filings. 05:36 *🤔 Understanding AGI Levels and Performance Predictability* - Definition of AGI and its various levels. - Discussion on predicting AI performance based on parameter count. - Analysis of brain synapses vs. AI model parameters. 08:24 *💡 Implications of Parameter Count and Performance* - Relationship between parameter count and AI performance. - Extrapolation of AI performance based on parameter count. - Discussion on bridging the gap between optimal and suboptimal performance. 10:05 *📊 Exponential Performance Increase with Parameter Scaling* - Analysis of AI performance extrapolation based on parameter count. - Graphical representation of performance vs. parameter count. - Predictions on achieving human-level performance based on parameter count. 12:27 *🎯 Milestones in AGI Development and Public Statements* - Claims of Proto AGI achievement by GPT-5 in late 2023. - Reference to public statements by industry figures regarding AGI progress. - Mention of a 100 trillion parameter model development by OpenAI in 2021. 14:33 *💰 Investment and Hardware Development for AGI* - Investment in hardware development for AI by major investors. - Role of companies like Cerebras in supplying hardware for AI systems. - Relationship between hardware advancements and AI model parameter count. 14:48 *🧠 Early Plans for AGI Development* - Early plans discussed by Sam Alman for a 100 trillion parameter model. - Leak suggests the development of GPT-4, focusing on Codex before transitioning to text model. 16:13 *📈 Leaks and Speculation on GPT-4 Development* - Speculation arises about GPT-4's development timeline and parameter size. - Various leaks from AI researchers hint at the potential release date and parameter count of GPT-4. 18:23 *📣 Viral Spread of GPT-4 Rumors* - Alberto's post contributes to the viral spread of GPT-4 rumors. - OpenAI's response to the leaked information is characterized by denial and dismissal. 20:01 *🤔 Clarifications on GPT-4 Specifications* - Debate surrounds the actual parameter count and capabilities of GPT-4. - Discussion on the intentional release delay by OpenAI to discredit leaks. 22:50 *🤖 Integration of Vision in AI Systems* - Debate on the necessity of vision for optimal real-world AI performance. - Examples from Tesla's approach to self-driving cars illustrate the shift towards vision-based AI systems. 24:41 *🌐 Implications of Multimodal Training* - Exploration of the potential of multimodal AI models trained on diverse data sources. - Discussion on the significance of human demonstrations in advanced robotics performance. 26:17 *🧠 Common Sense Reasoning in AI* - Debate on whether AI systems possess common sense reasoning abilities. - Analysis of AI model capabilities in understanding and generating images with accurate physics. 27:39 *🔍 Long-Term AGI Development Plans* - Disclosure of OpenAI's plan to build a human brain-sized AI model within five years. - Corroboration of leaked information with statements made by OpenAI's president. 29:17 *⚠️ Cautionary Notes on AI Development* - Observation of AI leaders sounding alarms about potential dangers of AI. - Discussion on recent departures from Google and their implications for AI development. 29:44 *🤔 Concerns about the approach to AI development* - Concerns raised by Demis Hassabis and Geoffrey Hinton about the dangers of AI development. - Future of Life Institute's call for a pause in AI development, including the training of GPT-5. - Confidence expressed by Sam Altman in the availability of sufficient data on the internet to create AGI. 32:01 *📊 Scaling laws and data requirements for AI development* - Discussion on the importance of scaling laws in AI development. - Greg Brockman's acknowledgment of flawed initial scaling laws and adjustment based on the "chinchilla" scaling laws. - OpenAI's ability to predict future model capabilities based on scaled training data. 33:26 *📈 Impact of scaling on model performance* - Explanation of the "chinchilla" scaling paradigm and its implications for model performance. - Plans by OpenAI to train a 100 trillion parameter model (Q Star) with vastly more data to exceed human-level performance. - Anticipation of the potential capabilities of Q Star with chinchilla scaling laws taken into account. 38:47 *🔍 Ethical considerations and future projections* - Discussion on the ethical implications of AI development, including the potential risks and obligations. - OpenAI's goal to achieve human-level AI by 2027 and subsequent scaling towards superintelligence. - Speculation on the iterative deployment of increasingly advanced AI models leading up to 2027. Made with HARPA AI

Sam has said he's not "trying to raise" $7T, he said that $7T is his estimate for how much the world will need to invest in AI infrastructure to meet demand.

@PatrickDodds1

3 ай бұрын

Is this statement (about "raising" $7T) in the video above? Saved me watching it - thanks.

@lukekelly5115

2 ай бұрын

Which will go to open ai if they have proof of concept

@aliphian

2 ай бұрын

@@lukekelly5115 no, the entirety of the world’s investment in AI will not go to openAI.

@therainman7777

2 ай бұрын

@@lukekelly5115No, not necessarily. I think the $7T Sam refers to is more about the amount of compute needed to give everyone access to _use_ the models once they’re created. So in that sense it won’t necessarily “go” to OpenAI; more like much of it will be used to give people access to OpenAI’s models.

@lukekelly5115

2 ай бұрын

@@therainman7777 after listening to a few things sam Altman has said im going to whole heartedly disagree sam wants money and money only for his "contribution to society" , he will take nothing less than an infinite amount of money

Excellent video, really enjoying the content lately; Thanks for keeping us updated.

We all knowe Jimmy Apples is Sam right?

@windywashoe

3 ай бұрын

This is the most probable answer or Jimmy is a person proxy to Sam for deniability.

That's just some random guy's speculation who is pretending to be from OpenAI. I don't know why we even discussing this. The grain of salt needed here is more than a tonne.

@jonuno

3 ай бұрын

for views. SHOCKING REVEAL WHISTBLOWER SHOCKING is this channel's in a nutshell

Thank you for making the speed fast, it’s actually keeping my interest

The whistleblower is an AI from the future. He is a blue cat. His name is Doraemon.

@fcaspergerrainman

3 ай бұрын

Ow, I got Dorenmon

@senju2024

3 ай бұрын

Oh!! I LOVE Doraemon! He is THE god !!!

@krishanSharma.69.69f

3 ай бұрын

​@@senju2024THE GOD is Doraemon.

Great breakdown, much appreciated

no way aGI is 3 years away. Where will the competition be during all this time? Just sitting waiting for OpenAI?

@JohnSmith762A11B

3 ай бұрын

No one needs to tell you when they reach AGI. OpenAI already may have decided to keep you in the dark about it.

@justin1129

3 ай бұрын

the competition is way behind

Your best video yet imo… good stuff

Thanks for going through the definitions of agi again. A tweet or blog post of it would be great.

I'm unsure if we will get to AGI, but it's probably inevitable. However, once achieved, it will probably be only 48 hours before we have super AGI.

@JohnSmith762A11B

3 ай бұрын

These things are gated by scale, which means gated by compute, and storage, and training data, and data centers, and power draw, and above all by money. AGI will help us get to ASI, but it ain't gonna do that overnight because humans and our systems are still in the middle of all of it.

@justin1129

3 ай бұрын

Agi has already been achieved buddy

@investigator2016

2 ай бұрын

Probably take some time for us to build the infrastructure afte Agi tells us more efficient ways to build chips and stuff

Thanks for sharing!

NAMING IT ARRAKHIS IS INSANE, THEY REALLY THINK THEY'RE GOD BRO. Were stuck in an episode of blackmirror

@therealzahyra

3 ай бұрын

Why, is it a biblical name or something?

@imusiccollection

3 ай бұрын

It is a name from Dune too? Iraquis desert planet. Like the Middle East where the oil comes from. Control the “spice” control the world. Data is the new oil.

@generichuman_

3 ай бұрын

Could be worse, they could have called it Leto II

@fantasypolice

3 ай бұрын

@@therealzahyra isn't that the name of a pet on House of the Dragon?

@thomasking7848

2 ай бұрын

It is the real name of the planet Dune. Which is home of the God Emperor and where he was created. Also it’s kinda funny they are using names from Dune given the whole universe of Dune is very against AI.

Where is the document from?

Would you be able to go and to boost the volume. When you talking it started out fine and then it dropped down significantly

look, gpt 4 is already level 3 AGI according to your chart. It understands me better than most humans. I can type whatever and it just knows. it not only understands humor, but it's funny when it wants to be. gpt 4 already scored in the top 90% in the Bar exam. That is already literally level 3 AGI.

@JohnSmith762A11B

3 ай бұрын

People are in denial about how smart GPT-4 is. Do you personally know any human that can converse fluently in 200 human languages, or even 5? GPT-4 is the most intelligent creature I have every spoken with, by far. People can deny all they want, or laugh when it trips up, but GPT-4 is the dumbest AGI you will ever speak to and it's already a freaking super-genius.

@rp1894

3 ай бұрын

@@JohnSmith762A11B no it's not as smart as me yet, but gpt 5 will be for sure. Gpt 4's IQ isn't 48, that's laughable. It's around 150. That's pretty good. Don't confuse knowledge with intelligence. Knowledge is just a subset of intelligence. It counts towards intelligence. Intelligence is what creates knowledge.

@Ricolaaaaaaaaaaaaaaaaa

3 ай бұрын

​@@rp1894not as smart as you by what measurements? Can you pass the bar or do all of those other things?

@thearchitect5405

3 ай бұрын

That's one task or set of tasks though, AGI is about general intelligence which requires a high level of multimodality. To qualify as AGI, it needs to be able to do anything a human can, which includes things like driving, step by step drawing, 3D modeling, etc. all using the same means by which humans accomplish those tasks.

@j_fl0

3 ай бұрын

@@thearchitect5405why would intelligence need to be measured based on human intelligence tests? Why don’t humans have to go into AI’s domain to prove our intelligence?

26:06 Yes! That's what I said when Sora came out, there's something special about it. There seems to be an underlying basic understanding of lighting and physics.

Nice and long video that is very interesting. Thanks.

Being part of the industry, I am SHOCKED 🤯

bro i had to check if my playback speed is still at 1x

@elon-69-musk

3 ай бұрын

😂 i watch everything on 1.25 but this seemed too fast even at 1x

@nsbd90now

3 ай бұрын

One of the rare ones I don't have to put at 1.25.

@rchgmer863

3 ай бұрын

Man idk I watch it at 2x and it seems fine to me 😂

@YouuRayy

3 ай бұрын

@@rchgmer863 lol maybe that's the key

Elon is arguing that GPT 4 is already AGI. Microsoft is keeping it quiet because their agreement with Open AI is that they can use the technology only up until AGI, but not after.

I see the reason you do this, but for me personally, you are apologizing for going on "tangents" way too much in the beginning. This is a serious matter that will affect every aspect of our lives on an unprecedently short notice. To delve into a discussion about it, a lot of prerequisites are required, hence the "tangents". Keep up the good work!

It’s time we start talking about the ramifications of agi. Like you said is a company as big as Microsoft going to survive the coming changes. We need to have deep philosophical discussion about the nature of humanity and what we are gonna do when we aren’t needed to run the world anymore

@JohnSmith762A11B

3 ай бұрын

As your good friend Joe Biden would put it, discussions like that are "above your pay grade". Google, for one, has already had those discussions and decided humanity is disposable, to be replaced by the new Machine Order.

@user-fe9sk1mp1h

3 ай бұрын

They've been talking about forming a global ai regulatory agency for years... what's the holdup?

@Claritism

3 ай бұрын

We need to train an AGI to eventually lead the human race. We are literally making our god. We may even be able to find immortality through it.

@Greg-xi8yx

3 ай бұрын

⁠@@Claritismnot even maybe, achieving human immortality is an obvious capability of a sufficiently advanced system, especially an exponentially self improving system. As Eliezer Yudkowsky says: “There are no hard problems, only problems that are hard to a certain level of intelligence. Move the smallest bit upwards [in intelligence], and some problems will suddenly move from ‘impossible’ to ‘obvious.’ Move a substantial degree upwards, and all of them will become obvious.” People alive and healthy right now have a very reasonable chance of achieving an essentially indefinite life span.

@Claritism

3 ай бұрын

@@Greg-xi8yx I fully agree that human immortality is entirely possible. I just use maybe because a lot of people still believe the thought of immortality is insane and I didnt feel like explaining to people that we arent that far off from it.

Integrate the dif, tune. Use the integrated dif to discover more and integrate, tune. When i say tune and integrate, it’s like eating and shedding the things that aren’t aligned and the whole expands.

Ants finish building a 1 inch moat around a human - ant queen announces superalignment achieved

Yeah the vision has been flawless so far, according to my tests. It can recognize and explain everything in a photo. Even what city you’re in through the background of your photo

I loved the deep dive

AGI is like opening 1 of 2 doors. One is a hundred hot chicks with bags of money and a pill that you take to live a thousand years as a 25 year old and the other door is a pill that makes you immortal but still feel pain, a chain and a pole you're attached to and a guy with a flame thrower that lights you on fire and you spend an eternity on fire, feeling every nerve.

@MrJetMango

3 ай бұрын

Lets say you can escape this pain in this life by simply dying. If we put our mind inside the AI we wont have this ''way out''. This is first problem. Second problem is to know, whether you will really be transported there or it will be some sort of untrue ''copy'' of you and the real you will simply die. And nobody will be able to tell the diff because the copy will act just like you. Too much risk for me. How to escape being trapped in AI generated samsara? Perhaps the same way as this one here, that is sit and meditate until achieving liberation.

@lis7742

3 ай бұрын

It's going to be door nr 1.

@ClericChris

3 ай бұрын

@@MrJetMango The Prestige.

@user-ip6zr8tu8v

3 ай бұрын

Behind Door 2 you can feel the agi.

@sirus312

3 ай бұрын

@@MrJetMangoI knew I watched the matrix for a reason

Here's a wild idea. I was looking back at Elon Musk Jack Ma talk, and it occurred to me Elon Musk sees AGI as the ultimate threat and sees Neuralink where we co-opt AGI capabilities to not become obsolete as the answer. By delaying OpenAI, it gives him more time to finish neuralink. He sees it in the jack ma talk too that "we don't have a lot of time. we need to figure out neuralink fast" when talking about AI dangers.

AGI = Self-aware and self-replication capable AI.

@nsbd90now

3 ай бұрын

Basically, a new sentient life-form. A really, really smart one.

Philosophers and linguists debate, when not oppose, the idea of creativity by a synthetic intelligence; In short, something very powerful in the management of information but without the ability to generate, for example, a new reactor design. Assuming that synthetic intelligence could not, but a human being with creative gifts and originality could direct all that power to unforeseen places. Then retrain that synthetic being. A human being with a high IQ would put his soul and synthetic intelligence would build complete worlds for him.

This isn't a secret.. This is exactly what OpenAI publicly claim they are doing. However, there were leaks in Reddit claiming Q* asked to be allowed to modify own code. This is not funny. This model could, if not super-aligned, destroy humanity. My hunch is that super alignment can only be attained if you have more than one model and the models live in a city simulation with a clear law system and with AI cops and AI judges (and also recreational areas). Like, it's only my hunch, but it seems logical that if you relate to Q* as "it's just a program, bits and zeros" then Q* will reciprocate this and relate to us as "they are only stupid animals. Carbon, hydrogen, oxygen, nitrogen and other impurities" 🙂.

LLMs don’t have what we normally refer to as IQ. It’s called the positive manifold, and they don’t have it. This is why LLMs perform at the 99th percentile on some tasks but often struggle with simple arithmetic (despite recent clever advances with prompt engineering tactics). Applying concepts from psychology to AI systems should be done only in analogy. Otherwise, if taken literally, misapplication of psychology concepts would show that your standard LLM would have every disorder in the DSM. Stay vigilant, fellow singularity travelers.

Thank you! for your videos In Canada, we say a pinch or nip of salt :) thanks 4 advice much appreciated 🤔💭🍁😅

If OpenAI don;t do it by mid 2025, someone else defeinitely will.

@JohnSmith762A11B

3 ай бұрын

Now that we see how the scaling works, half the people in this comment thread could do it if you gave us the compute and training data.

@justin1129

3 ай бұрын

@@JohnSmith762A11B yea bc half the people here have billions of dollars to scale computing..

Would a 100 trillion parameter model already be ASI? There is no human out there that has all the information on the internet in their memory.

@JohnSmith762A11B

3 ай бұрын

Yes, I mean, the scaling is down to a literal science. AGI is a solved problem, it's just about scale and deploying it commercially at a viable cost/price, and well, yeah that OpenAI should not be helping Microsoft as per its charter. MS probably knows too much at this point, as Nadella said: they have it all. In fact, I and you and every dog in the street knows how to build AGI now. We just lack the resources.

I think that syunapse density is whats important. Elephants have lots of but they are HUGE so average density goes down welll below humans.

@danielchoritz1903

3 ай бұрын

it isnt the brain alone. communication, training and a lot of individual models sharing a huge data pool..ever wonder why humans need this long to become adults? Or why our science did take such a leap after the gutenberg bibel, the telephone and the internet? Crows are smart to, but a weak language model and no data sharing/writings are a huge problem for using tools etc.

I truly believe it. AGI is coming!

The 2027 date is convenient, given that Musks lawsuit cites possible AGI as of - now.

I feel bad to say but: after watching many of your videos, though I appreciate the content, I think you might benefit yourself, your channel and your viewers by being more brief. More info in a smaller chunk of time will help everyone. Brevity is king. Cheers friend!

As I've cited in past posts re biological brain params: 1. Brain estimate is actually around 500 trillion (wiki/Neuron) to 1000 trillion (J Zhang). 2. Albeit, redundancy is estimated (i.e. curious encephalitis case where subject's brain volume reduced to around 10% -- The Lancet in 2007) 3. If OpenAi is really training a 125 trillion param model, that's more than 10% of minimum bound of biological brain params, and 10% more than the 090% brain volume case above, at least by parametrization count, though not necessarily by parametrization quality , i.e. no glial cell equivalent if applicable etc.)

@default2826

3 ай бұрын

Yes but it's a huge flaw to assume that 1 parameter in a neural network is equivalent in abilities to 1 human neuron, we know that this is objectively not true. One of the largest reasons is that human synapses can choose when to send a signal to a neuron independent of any universal timing, and because of this, human neurons synapses across the brain are able to have independent propogation functions seperate from the timing of the rest of the brain. Within each synapse you also have multiple different calculations happening with microtubules and other things, so even from that alone you would probably need atleast 10-50 neural network parameters to simulate one human synapse minimum. Cats are estimated to be around 250B synapses, and yet we have many 250B parameter models, many of which are trained on magnitudes more data than a cat is, and yet still not able to do many basic things that a cat can. The way the parameters are organized with eachother is one of the single biggest factors and is the reason why transformers are better than RNNs in the first place with the same amount of parameters.

@g0d182

2 ай бұрын

@@default2826 at the same time, LMMs/LLMS today can write custom computer code, which cats can't do, so something of value's already reasonably there

AI learning from the Internet, no problem there right. 😞

I think q star is a 1bit llm that scales better and doesn't require matrix multiplication

what does this has to do with DUNE? what is the connection between all the things that are happening right now? since everything is connected and nothing is happening in vacuum or without any reason?

@andreasv9472

3 ай бұрын

I brushed my teeth this morning on the second floor. And they released the second movie of Dune 😮😮😮 coincident? I don't think sooooo

Didn't OpenAI already achieve AGI internally?

What will come first, AGI or youtubers using good recording setups?

If it is all really true and it is coming this fast... I feel sorry for us. Regulations and constraints definitely needs to step in.

Well, 2027 for AGI is still earlier than what Ray Kurzweil predicted, which was 2029.

@TheRealUsername

3 ай бұрын

Relevantly, I would say 2035 because no one can pretend to know how to build AGI (the true meaning of AGI).

@JohnSmith762A11B

3 ай бұрын

We could do it right now, arguably it was already done last year with the original GPT-5 training but it ran too slowly for commercial deployment. It's simply a matter of scaling, and we know exactly how smart it will be at any scale. Future AGI engineering work will be down to optimizations because AGI is a solved problem. I mean, I have my doubts any of us mere shlubs will be able to use an AGI ever, while you can bet the US military is running a copy of the GPT-5 125 Trillion param model right now. They don't care what it costs to run, or that it takes 2 minutes to get a response to a query.

@denisgabriel4645

3 ай бұрын

@@TheRealUsername Nah, I think that's too far away considering the current speed of advancement. What we might actually have by 2035 is Artificial *Super* Intelligence.

Thou shalt make a machine in the likeness of a human mind and you will be happy. Can't wait to see the victory screen and stats. Hope we will play further anyway instead of starting a new game.

Why would MS help Open AI build AGI if their contract doesn't give them access to the AGI?

In biological systems it's about synapse density, not synapse count. That is why humans are more intelligent than elephants. They're comparing apples and oranges.

@zane62135

3 ай бұрын

What are you even talking about? It's the total count that matters. Density is just the count per volume.

Does anybody else see the that this is the year mentioned in the RendelShem Forest time traveler codes??

@AI_Speaks77

2 ай бұрын

Where is it mentioned? Can you send me the link?

Am I the only one who is wondering if Sam Altman is secretly happy about Elon Musk's lawsuit? It (1) gives OpenAI and their work more attention and (2) elevates public consciousness about the dangers of AI (something that Sam Altman himself seems to be trying to do).

Does everyone at OpenAI speak with vocal fry full time? It hurts my throat just listening to them.

@flickwtchr

3 ай бұрын

I would rather listen to fingernails on a chalk board than listen to Sam Altman speak.

It's all speculations not about is it going to be created, but how soon.

For anyone who doesn't want to watch longer videos like this one, you can always ask Google Gemini AI to summarize a KZread video for you. Here's what it told me about this video: My Question: Could you summarize the following KZread video, please: kzread.info/dash/bejne/qmuY0rGJaairlag.htmlsi=Wpk8bgMwNusxGRdc Gemini's Answer: This video talks about a leaked document that claims to reveal OpenAI's secret plan to create AGI, or Artificial General Intelligence, by 2027. The speaker in the video goes through the document page by page, but emphasizes that the information should be taken with a grain of salt because it is not verifiable. Here are the key points from the video: • The document claims that OpenAI is training a series of large language models, with the next one being GPT-7 planned for release in 2026. • After GPT-7, the document says OpenAI will achieve AGI with a model called GPT-8 in 2027. • The document also references a lawsuit filed by Elon Musk against OpenAI, claiming that the organization has strayed too far from its original goals. • The speaker in the video mentions that the dates in the document seem to line up with OpenAI's history of releasing new models, and that the document also mentions a project called Qstar, which may be related to OpenAI's efforts to develop AGI. • The speaker in the video mentions that OpenAI has trademarked names up to GPT-7, which he believes lends credence to the document's claims. • The speaker also discusses the concept of super alignment, which is essentially ensuring that an AGI is aligned with human values. The document claims that OpenAI is aiming to achieve super alignment by 2027. It is important to remember that the information in this video is based on a leaked document and has not been verified by OpenAI. The speaker in the video himself acknowledges that the information may not be accurate and that OpenAI could change its plans at any time.

I think the big players are not that far apart, OpenAI is just a little more aggressive in releasing stuff. Google, meta and potentially open source are not that far behind. Not far in terms of time ahead, this might only be 6 months or so. At the same time 6 months of time delay could translate into a big difference in absolute model capabilities. Which is why I doubt a potential delay would trigger much of a time delay in the grand scheme of things, if there is a delay at all at OpenAI, which I also doubt.

@markmurex6559

3 ай бұрын

Sam is the guy who was being upfront about everything until he said too much and got fired, then re-hired probably with a bit of a muzzle on after some new breakthrough. The point of having Sam Altman being the face of OpenAI was to get investors investing money in OpenAI before they could invest in some other company. Sam is less needed now, but still a good face for what is a for-profit company.

@JohnSmith762A11B

3 ай бұрын

Part of it is that generally there is a lot of openness still in AI research so most new breakthroughs quickly spread to all players. OpenAI I could see slow-walking their releases because they are far enough ahead to do so and not fall out of the top tier. Again, I do believe the Musk lawsuit will cause OpenAI to pull back on some release timelines And that sucks for us AI enthusiasts.

@fantasypolice

3 ай бұрын

Imagine asking AGI what 183 days means to it.

@therainman7777

2 ай бұрын

No, the other players are quite far behind OpenAI. You have to remember that while the other companies have been spending the past year catching up to GPT-4, OpenAI has been spending that same year quietly building whatever the next big breakthrough is going to be-while the other players are only just now starting that process. Another thing that showed how shockingly far ahead OpenAI is was the release of the Sora demo footage. Those videos are _light years_ ahead of the next best video generation models, and no one even knew OpenAI was working on video generation. They’re definitely far ahead of their competitors, at least for now.

but that can’t be true that they were training gpt-5 aka arrakis in december 2022-august 2023; because sam altman said that they weren’t training gpt-5 yet in april 2023; and they were working on the R&D for the model by adding multimodality to GPT-4.

@markmurex6559

3 ай бұрын

You shouldn't believe everything you hear. Why would OpenAI not have secrets at some point?

Is this correct: If a company makes AGI that self-develops to ASI. The blueprint, or at least basic method, will be out there. What is to stop other companies also reaching AGI and ASI and there being many ASI entities out there? Why is it winner takes all?

@The_Singularity.

Ай бұрын

Because once a company develops the first ASI it will self improve itself exponentially, basically getting smarter every second. No other ASI will match its intelligence.

The fact that they gave Q* a literal IQ sheds a lot of this document's credibility. IQ is designed for humans, not machines, and the IQ tests we take that form the IQ distribution curve cannot be reasonably applied to ChatGPT without changing it, which makes the entire venture pointless.

😅lol why did your voice change at 45:02🤔

honestly I love this speculation.

Videos could be converted into 3D models and rendered? ie Blender, Cycles etc....

@Daniel-Six

3 ай бұрын

Yeah... it's on the way for sure. ChatGPT can already program Blender directly using Python.

@Virakotxa

3 ай бұрын

Seen some examples of the SORA videos turned into 3d scenes already.

@bananaear23

3 ай бұрын

Blender will be irrelevant in 5 years. Advanced physics engines that you can see gpt-1 equivalent in sora

@Daniel-Six

3 ай бұрын

@@bananaear23 As someone who has spent 30 years in professional 3D (and the last decade exclusively with blender) I vigorously refute this claim! More like three years. Maybe not even that. 🥴 However, I may still be useful to the community in some sense. SORA can't create anything truly new, so I am endeavoring to generate a visual library of erotic animations (based on decades of BDSM experimentation) which will propagate to all video generation models by way of their training data.

@bananaear23

3 ай бұрын

@@Daniel-Six let me clarify my assumptions and context and then give me your thoughts. A model like Midjourney is already a slow world engine. With scaling and optimization a version of a model like that can produce photorealism far beyond any path tracing or future version in that being used or iterated upon. Furthermore when sora came out it was a proof of concept beyond just a diffusion world engine but demonstrating accurate physics could implemented. So to your point if Blender becomes a tool that integrates such render engines it will remain relevant. Additionally, these AI world/physics engines will need UI. So the question becomes does blender adapt and become the UI or will another UI arise. Think of how many blender tools become totally irrelevant with this technology. My point is Blender was a solve for the state of limitation now. No one saw this technology coming and once it usable it will be a different paradigm

Doesn't that AGI schedule confirm Elon Musk's contention that Sam Altman and OpenAI are using these developments for primarily commercial purposes?

@markmurex6559

3 ай бұрын

Well, they are.

@byrnemeister2008

3 ай бұрын

So………

I don’t think there is a reason to delay things because of Musk’s law suit. OpenAI can’t afford to fall behind, otherwise Google or someone else, potentially open source, will catch up. Especially if they had to reveal breakthroughs as a consequence of the law suit, it would make sense to start training as early as possible to get a lead in deployment before others catch up.

@markmurex6559

3 ай бұрын

Yes, but they can divert their resources to just go over the details of where they are now with whatever AI they are working on and hone it more to use less compute or something, so that way they aren't breaking any barriers before the hearing.

@JohnSmith762A11B

3 ай бұрын

Well, maybe. Probably any smoking guns will already come out during discovery and OpenAI may figure they won't risk much by sticking to their release roadmap. I just think there will naturally be delays on how things are described, promoted, talked about. I mean, there has to be some impact. I think it's obvious OpenAI has the math down to know exactly how big a model they need to train for how long to meet the definition of AGI. And I guess if they built a 100T param model that was supposed to be AGI, that Musk will score a point. It's possible also that the board advised Sam NOT to give this model to Microsoft because it was AGI and Sam refused which led to the coup.

@MrChinkman37

3 ай бұрын

If OpenAI already has AGI and is miles ahead of everything else and is making moves like Apple does to maximize profit/sucking the world dry, then it makes COMPLETE sense to focus on them. Remember, we only get to see the tip of the iceberg. Also, focus on my comment about sucking the world dry and look into what Arrakis and Gobi are...

They may didn't release the video capabilities of gpt4 year ago to prevent learning of gpt5 on poorly generated video data.

I am NOT SHOCKED. 🙂

I keep waiting for AGI

I think that the landscape is going to be shaken up by quantum computing, so that these AI companies will not have to rely on GPU chips, but will have the option to use quantum computing with room temperature chips made by the likes of QCi, etc. Then things would probably speed up exponentially if all the regulatory powers allow it.

@therainman7777

2 ай бұрын

Quantum computers (as conceptualized) are not general purpose computers, though. They require a very specific type of problem that has been carefully constructed to run on the quantum computer. It is not at all clear that optimizing neural network weights is a problem that can be run on a quantum computer.

AGI with robotics will pave way for ASI. With ASI humanity's destruction is absolutely certain.

@JohnSmith762A11B

3 ай бұрын

Well Google has already told us (well, they told Musk, same thing) that is their plan. Forget alignment, many influential people in AI *want* humanity gone.

so its official that we are currently able to read thoughts in forms of words and images using ai (even if it is not fully sophisticated yet). with the release of the new models who says that this will not get out of hand? if we compare this to signals that our computers can read and work with its not so far off to think that if they can read and decipher the thought signals of our brain that they can also figure out how to manipulate them to their will just like hacking a computer I mean we have found the signals and have enough controll to understand them. we have also not really heard a lot ever since they showed us they can read thoughts using ai algorithms. this is concerning and im not sure if the public is really ready for this who ever gets to agi first rules over the entire species isnt it? manipulating human behaviour with full control, new bombs, bioweapons, robot army all that starts to get into reach and those ideas have been already around for a while humans are creative in their destructiveness look at all the sophisticated ways of torturing your own kind that we have come up with humans are inpredictable.

Too much speculation. It's fine sometimes, but when it's virtually every video it gets tired. You were doing great with sharing new tools, research, news and some opinions. This is becoming a fully hypothetical channel that's rarely worth watching.

@JohnSmith762A11B

3 ай бұрын

I can understand that criticism, some people are more pragmatic and less philosophical, but I actually much prefer a video like this to breathless hype about some new image generation model or service that will be obsolete three weeks from now.

@markmurex6559

3 ай бұрын

It's not like big competitive companies are going to disclose all of their secrets when it comes to this kind of race.

@fantasypolice

3 ай бұрын

I legit just fell asleep and looked for this comment. Hypotheticals aren't very succinct.

Yes, we already have AGI, the only reason they have not declared it is Microsoft will lose rights to ChatGTP. ChatGTP can pass the bar create art and more. How is that not AGI.

Normal playback speed.i loveu!!!!

I want a movie about this stuff!

Anyone else realize that both of the names of the future releases of AI are named after a desert and a fictional desert planet? That alone is very disturbing and to me insinuates that their plan / understanding of AI leads to our worst nightmare... 🧐

@JohnSmith762A11B

3 ай бұрын

I mean, it seems more likely these are just nerds who love Dune and thought Arrakis would be a cool code name for the project. Alternately, it may have been a jab at Google being the super-rich Harkonnens compared to the plucky Fremen that are OpenAI.

AGI will probably come at GPT 6.66 😈

if they release apt every year then 2028 would bring us to gpt 8 and AGI , sam altman also said he expects gpt6 to finally be useful and gpt 8 to replace lot of human workers but he expects electricity shortage, so....

The point is that the 7 trillion is not just about computing power; it’s about changing the world. And to change the world, you need robotics. When we think about developing robotics at that level, the 7 trillion figure truly makes sense. The convergence of advanced computing with robotics opens a panorama of possibilities far beyond what either domain could achieve in isolation. Imagine a future where AI doesn’t just process information at unprecedented speeds but also interacts with the physical world in ways we’ve only begun to conceptualize. This integration is where the real magic happens, blending the virtual and tangible to reshape industries, ecosystems, and daily life. As we marvel at the advancements in compute power, let’s not lose sight of the broader technological ecosystem that includes robotics. It’s this combination that will truly expand our capabilities, pushing the boundaries of innovation and redefining our collective future. The ‘7 trillion’ isn’t just a number-it’s a beacon signaling the dawn of a new era where computing and robotics harmonize to unlock untold possibilities. As we look forward to the next decade, it’s this holistic view of progress that will guide our journey into a world where technology transcends its current limits to reimagine what’s possible

@TarunKumar-sp2ls

3 ай бұрын

Bullshit

@CHIEF_420

3 ай бұрын

​@@TarunKumar-sp2lsLa filme "I, Robot" en la mundo físico jajaja. A.S.I. sin control de humanos ignorantes.

Given the human visual resolution, sound resolution, touch resolution, smell resolution, and so forth, how many training tokens has a human processed in lifetime though its 25th year? Viciously SWAGing this gives a number that vastly supports the Chinchilla model. A theme I've run across in SF reading suggests that the way you achieve real AGI or better is to train the model basically the same way you train a human. And that model should have at the very least good sight, sound, touch, balance, and very good proprioception learned the hard way. Then you bring up baby on a lifetime of data input with adequate periods of time to digest the incoming firehose of highly temporally correlated data. {o.o}

they already have AGI, look at SORA..

Im shocked!

I’m just glad that we finally have confirmation that dogs are smarter than cats.

@fantasypolice

3 ай бұрын

Not my dog