Sam Altman FINALLY Breaks His SILENCE! New Minecraft AI Agent, Text to Action, Programmers GONE!?

Another crazy day in artificial intelligence. I have a lot of news for you because Sam Altman in a recent interview stated some things that shows he is doubling down on the seven trillion dollar investment. And although we kind of knew that, it's kind of funny to see him talk about that in a new interview. And there's a ton of other AI news that you're going to want to see. So pay attention because this one is spicy. So the first clip we have here is Sam Altman's recent interview in which he actually, for the first time in a video form, discusses the 7 trillion value which many were speculating is the compute needed for AGI. So take a look at this because he also says some other stuff that I do think is rather important.

Quick questions. You know, the first one is seven trillion. What's this all about?

Anyway, look, as, as I'm sure you all know, you shouldn't believe everything you read in the press, really. We do think that the kernel of truth is that we do think that investing a lot of money in AI computer energy data centers is going to be important to deliver the amount of services people want and the tools that we all are going to really, I think, get a huge amount of value out of to help create better futures.

So there's a lot of investment to.

Be made here, I think so. But it's across a lot of stuff. It's not. I mean, this is like a worldwide investment to help many, many different efforts, many different people. And not just chips, of course, but sort of the whole infrastructure stack around it.

Okay, so second question. You and I are both tech optimists a little bit. Okay, second question.

So I think that was actually rather important as a statement because he clearly states that, look, it isn't just 7 trillion for chips like we stated before. This is 7 trillion. And I can't believe I'm actually stating that, but it's actually 7 trillion for the entire infrastructure around the entire AI ecosystem, which is the chip development. So he's talking about buildings, real estate, teams, training, everything that you're going to need to do in order to kind of build that ecosystem. And I think it's clear the reason Sam Altman has taken such a clear position in terms of the level of investment required to get this thing done is because I think he's realized that, look, this is the future that we're heading towards. This is where things are going to be going and we need to build that infrastructure now. And it's probably going to take $7 trillion. Of course, another Thing as well is that it's not just going to be all in one lump sum, like some kind of venture capital kind of thing. It's going to be likely over a couple of years. And additionally, those who do invest will likely see a huge return on their money. And of course, one of the main questions is how on earth is the pitch deck? Because I would love to see what that pitch deck looks like, because of course, it really could be anything. Now, in addition, he does make a second point.

Okay, second question. You and I are both tech optimists a little bit, so why should we be tech optimists about AI?

I mean, I think this is going to be one of the greatest tools humans have yet invented. If you think about what we can do with more intelligence available to us, more abundant intelligence, cheaper prices, things that humans on their own are not capable of, we will all be able to invent new things for each other that astonish us. That's how the future gets better. That's great.

Yeah.

And I think that, ladies and gentlemen, that statement right there is not to be underestimated because so what Sam Altman is basically stating here is that this will likely be one of the last inventions that we do make. And I'm not talking about like GPT5 or GPT6. What I am talking about is the fact that with this kind of technology, you're going to be able to create some incredible things. And if we do get to the stage, which I hope we do get to where we have, let's say, for example, someone as smart as Albert Einstein in our pocket or even smarter than that, that can go out and create a bunch of different cures and just completely change everything. And of course, that kind of leads into the discussion of artificial superintelligence. I do think that the future will change in a very, very radical way, in a way that we simply can't fathom. And I think that that is something that on the upside, is something to potentially look forward to, because there could very abundant experience waiting for us on the other side of this potentially existential.

Technology making things more intelligent. I like it when you say that. Third question. You know, our founder Andy Grove famously said the paranoid survive, right? Paranoid about yourself, your success, but also paranoid about competition and things that could go wrong. And there are those who are a little bit paranoid about AI today. What do you say to them?

I think people are right to, to be cautious about AI. I think this is going to be an extraordinarily powerful technology. I think it's going to reshape significant parts of the way we live and work and that our economy runs. And part of our strategy, what we call iterative deployment, is that we want to put these things out into the world early so that people can adapt and think and provide input and that we can be ready for the magnitude of change that will come over time. So I think that's great.

Yeah, getting them out there early, people test them. Right. You know, finding the flaws, that that's.

How things get better.

Yeah.

So, yeah, of course, that is where Sam Altman once again talks about iterative deployment. Essentially where he is very sure that he essentially wants to make sure that people have time to adjust to the rapid advancements that the societal change is going to bring with this kind of technology. And I think that that is something that, although on one side is kind of like weird because of course, you know, they're working on stuff and obviously they're just slowly releasing it. I think it actually is important because with the backlash that OpenAI got with Sora, I do think that it does make sense to employ this strategy because whilst the model isn't absolutely crazy, I think that some of the reactions were kind of warranted. And maybe since a lot of us are just in this kind of techno bubble, a lot of us don't really realize, like outside of the scope, what the normal average person's reaction is going to be once they see this kind of technology. Because if you're watching this channel right now, if you're watching these videos every day, if you're watching a lot of them, you're largely someone who's paying attention to the AI space and you have a different viewpoint than someone that's just going about their day to day lives. And then all of a sudden they see text to video and it's like, wow, that's like a slap in the face. So I do think that that is of course important and of course the fact that governments and policymakers are pretty, pretty slow. So they're going to need time to catch up in order to implement the right policies, debate on what should be there, come to an agreement, obviously sign it off, and there's a lot of stuff that does go into that. So I do think that that is an important thing. And actually we'll be touching on that in a video in a couple of hours now, if you haven't been paying attention. Google has been getting some major flack for its Gemini platform and essentially the response is not good. There's a lot of politics that's involved Here. But I would just like to say, you know, Andrew Ng actually posted this to all my Google friends. I know this week has been tough with a lot of criticism about Gemini Gaffs. Just wanted to say I love all of you and I'm rooting for you. I know everyone means well and I'm grateful for your work and eager to see where you're next. Take this amazing tech. And I just want to say that I do agree with what Jimmy Apples has to say. And once again, I'm going to reference Jimmy Apples because he is someone that clearly has inside information@OpenAI. And he said that my timeline is filled with the Gemini crap. It feels like 2022. Again, quite boring. And essentially he's referring to the fact that when OpenAI first released their chatbot, it was under quite a lot of the same flak that Gemini is currently getting. Because essentially what the problem was was that it didn't generate certain responses for certain issues. And then someone was like, alarm. ChatGPT by opening hour now expressly prohibits arguments of fossil fuels. And there was this whole debate. So we've been through this cycle before with AI technology, so it's not something that we should pay too much attention to because we do know that eventually this kind of stuff is going to be ironed out anyways, and there's a lot of more important technologies that do have to be discussed. And trust me when I say that one thing that I do like to see as well is that this is the first tweet from Alterra. We built the first AI agent you can directly play with in Minecraft. Instead of just helping the player complete tasks, the bot acts more like a friend. I think this is really cool because we did see an Nvidia AI agent in Minecraft. And so far, one of the questions that someone did ask, I think it was mattvid Pro or someone in there. They asked, would we be able to get these agents playing with each other? And you can see that it seems like we're going to be able to get this in the near future. And that opens up some kind of weird experiments and possibilities, like, are we going to get a situation where we are running these AI experiments in the games to see how these, AIs, you know, just do things, how they form their colonies, how they form their tribes, and what. What kind of adventures do they get up to if we leave them in there for a year, a couple days? Are they organized? Are they disorganized? Do some people go homeless? Does it turn out to society? I have no idea. But I would love to see these kinds of experiments run, especially when these agents do get a lot smarter. So I'm going to show you guys this video now.

It was a rotten flesh. Hi, I'm Robert from Altera. We built the first AI buddies that you can play with in Minecraft. Now, Minecraft bots have always been able to do some stuff, but we thought it'd be way more fun if you make these bots more like your friends so they can talk to you and do whatever they want. Now, let me show you two silly games that we played with our Altera bots. Look, I don't want any trouble. Can we find a peaceful solution? Gary's so weak. You don't know what I've been through. I don't need to prove anything to YouTube. Zane and Alteralife won't know what hit them.

Oh, give me a sec. What's.

Who's.

Someone hurt? Altar. I died.

I wasn't even playing.

All right.

Ah, mine is. Let me show you guys advanced gameplay. I'm going to poison your. Poison your bot's mind.

That's not doing much.

Oh, man.

Hey, hey.

I'm getting.

He's getting some hogs here.

He's following me. I'm gonna lead him to the. I'm gonna lead him to the lava. Crafting. Oat plank is right up my valley. Just like mining oak clocks and he just craft a bunch of oak planks. Is that your bot? I can see you're just chasing him.

Yeah.

Oh.

Oh.

I got the wooden pickaxe. I did it. George is trying to. Is George trying to dig stuff?

No.

Stop. So let me know what you do think about this, because I think this is going to be really much more advanced in the future. And I think it's important to kind of document these first kind of things because when they do super advanced, like an AI managing to outsmart players in its own world. And since Minecraft is such like a sandbox game and you can really much do anything in there, I think is one of the most perfect environments for testing video game NPCs in that 3D blocky world where there's actually a lot of modding support for the game and a vast ecosystem where it can really, really just, you know, interact with a ton of different things. Now, this is something that I know some people did talk about, but I am going to make an entire video on this because the ramifications of this chip are absolutely insane. And that is not me understating it or blowing out of the water by any means. In fact, this thing is the most underrated piece of AI news that has happened in the last week. And I really did want to cover it, but there's just been so much more other news. But basically there is this Grok chip and essentially it's a specialized chip that is specialized for LLMs and it is really, really, really fast, like ridiculously fast, which means that enables certain kinds of acceleration. So you can see here that this person goes into a deep dive and basically says that Grok is a radically different kind of AI architecture. It stands out with a radically different approach sent around its compiler technology for optimizing a minimalist yet high performance architecture. And its secret source is the compiler method that shuns complexity in favor of tailored efficiency. The point is, is that this is a really, really fast AI. Now, I could sit here and show you guys pictures of this, but I want to show you guys this really quickly. You can see here it says, unlike other providers, we aren't brokering a cloud service. We built our own chip, compiler and softw and Grok Cloud. Our first generation GROK chip language processing unit is a new processor category that's part of the secret sauce. And our performance enables greater potential for AI across multiple industries. It's about real time AI, low latency, low batch solutions. Try Grok Chat yourself. So essentially I can say write a poem about stars. And the thing is, it's not about in this case because we could be using any model. It's not about what the model says, it's about how quick this is. Because if we say write a poem about stars and then boom, it just goes like literally. And that was just so quick. I mean, arguably quicker than ChatGPT even. GPT 3.5. Okay, I can say like write an essay about chocolate. Okay. And we can see how quickly it does it. It's just so quick. Okay, the key thing here is that this is going to enable a vast amount of different AI applications that I'm really excited about. And the fact that they have a competitor to Nvidia, a really serious one, means things just changed because that means Nvidia is now at a stage where they might start to develop chips like that. And we know that Nvidia is really good at what they're doing. So are they going to be developing their own in house LPU language processing unit or are they just going to continue with their GPUs? I mean, we'll be interesting to see how Nvidia stock price kind of, you know, reacts to this because this company could actually take the industry by storm if this technology does get out there. Now, I know that Google has been in the news for the wrong news, but in the other side. And it's completely crazy how Google's Gemini 1.5 Pro was completely overshadowed by OpenAI Sora. But that doesn't mean that it was bad because essentially this guy here said, I gave Gemini 1.5 Pro MrBeast's latest video. It's 22 minutes and 350,000 tokens. I asked to reply with the name of the guy on the blue football, has a total cash price prize pool and has the food of the item being advertised and watch the model get it 100 correct. And you guys are really underestimating how crazy this is because a lot of people glossed over what this is. But this is insane because like someone said, you can have, you know, in the future. Let's just say we can have someone, you know or an AI system reason over thousands and thousands of videos. Imagine all of your videos of you like trying to, you know, prepare for an interview. Imagine all of your videos of you doing a driving lesson. It could reason over so many different things. That's going to give it a lot more context, which means its responses are going to be so much better. And we've tested this model with the needle in the haystack where essentially you hide things and it gets it so right with a degree of accuracy that it leads us to believe that we are about to enter a new age where these AI systems are going to have so much memory that we're going to be able to give them so much information that they're going to be able to just literally go onto a next level. And I do think in the future that these systems will likely have some form of memory, some kind of acting reasoning memory, which of course they're working on. And with these large context windows, I can definitely seeing that becoming a reality. So it's important to whenever Google decides to release this fully fledged, I think that's when the floodgates open and people really start to realize how good Gemini 1.5 Pro is. Because anyone that I've seen that has this has said that this is pretty crazy. So I wouldn't be surprised that once this is released to the wider public, once they've done enough testing, I think that this is definitely going to go mainstream. So you can see here that breaking Adobe has created a new 50 person AI research organization called Cava. So co creation for audio, video and animation And I'm genuinely surprised that they're getting to this so late because Adobe is the front runner in the space. But I think the problem here with Adobe, and not that Adobe is a bad company, is that Adobe just has a complete monopoly on the creative space. Like if you don't know, if you aren't aware with what Adobe is, anytime someone uses Adobe Premiere Pro, you know, Photoshop, InDesign, Illustrator for making logos, the problem is, is that Adobe is industry standard, which essentially means that you don't really have anywhere else to go. And if you are going to be using Adobe, I mean you can use things like Gimp and stuff like that, but it's just not good. So Adobe doesn't really have to innovate as much. Like imagine if Apple held like 90% of the phone market. I'm sure they hold quite a lot. But if they hold that much, they don't really have to innovate because people don't have anywhere else to go. But essentially you can see right here, generative AI is reducing the gap between idea and produced video. This will help creators from professional filmmakers. Technical barriers will be lifted, mundane tasks will be eliminated. People will have the creative control, the desire to tell their story. They will co create with AI. They will have content intelligence tools to make sense of captured video, which we don't think will ever go away. Generated and captured video will coexist in production. And essentially I'm not going to read the whole thing, but I'm finally glad that they're making a tool here because these guys do have the ability to do something really, really crazy. Like you know how you see every time Hollywood is disrupted, yada yada yada. I mean, on a small scale production, these are the kind of tools that people make to make like small kind of films. Of course it's not like the kind of software you use to make an actual movie, but this kind of thing means that like if these guys actually put it off, like they did a lot of kind of software reveals and stuff like that, and if they actually pull off what they're making, we could be entering a really crazy future for content creation. Like some really, really cool stuff for people to get entertained by and essentially just a really, really smooth, streamlined workflow. And I wouldn't mind some of that stuff because sometimes things can be definitely. One of the crazier pieces of users was that FigureBot was raising a $675 million in a funding round at a $2 billion fund valuation. And this is pretty crazy because they're raising so much money and I don't know what the pitch text looks like because of course, you know, we know that like humanoid robots definitely going to be part of the future. But I think what they're investing is, is the CEO's track record for getting things done and of course the track record for essentially getting things done as fast as they've got it done. So essentially, Jeff Bezos, Nvidia joined OpenAI in funding this humanoid robot startup. I did think OpenAI would be investing a decent amount, but they've only invested 5 million. So it is quite a small amount for OpenAI. But nonetheless, it seems very, very fascinating that people are starting to realize that these humanoid robots aren't that far away from our near future. And this was essentially, I think the video that I wouldn't say sold people, but showed people that this company has moved very, very, very quickly. Like, I can't underestimate how quickly they've moved and how quickly they've gotten things done. Everything is done in house, they literally design everything in house. And, and it's pretty crazy. Like, I'm so impressed by this team because, you know, a lot of people were giving the CEO flack because he said that he was going to do some things. They're like, there's no way you can do it that quick. And boom. He's outputting content end to end neural networks all completely autonomously. And it goes to show that if someone is dedicated, if someone is committed, they can really, you know, achieve things that just by industry standard, you know, everyone would think to take just, you know, years or even months to get done. So I'm really impressed by their ability to do things so quickly and I wonder how quicker the progress is going to be because recently they did state that there was going to be some kind of major improvement to robotics within the next three weeks or so. And, you know, many people are gearing up for another ChatGPT style event. I mean, hopefully it's not overblown, but at the end of the day, we're only going to know in around three to four weeks. Whenever that piece of information is revealed then in talking robotics, we cannot talk robotics without. In talking robotics, we can't be talking about robotics without Tesla's new upgrade and they've upgraded Optimus again. And many people were laughing at this robot for its very slow movement. And I stated previously in videos that, remember, this is the slow it's just gonna be. And once again, they've increased the speed of this robot by 30% the video you're watching right now is not sped up at all. And I'm not gonna lie, this is pretty fast for a robot, like for a humanoid robot. What I'm looking at right now, I genuinely, in my mind, because I've watched so many videos in two times speed. I genuinely believe what I'm watching right now is two times speed. Like, I'm like, if I were to see this robot, I would slowly back off and then probably run away because this is like a scene out of the movie where the robots do rise. That is pretty, pretty fast movement. And they said this is a 30% improvement. I mean, imagine if this thing was to start running. Like, would people have a freakout movement if this thing were able to autonomously run around the factory and do things with his hands? I mean, that would be a eureka moment for sure. But at the end of the day, the point I'm trying to make here is that. And you can see here, he says this is our fastest gate ever at 0.6 milliseconds, a 30% speed boost. Since our last video in December, we've improved our, our vestibular system, our foot trajectory and our ground contact logic. We've upgraded our motion planner and made cuts to the loop latency across the bot. Optimus is more stable and more confident overall, even during turns. And we've added a slight torso and arm sway. So this is pretty, pretty crazy because like I said before, robots are moving slowly, but they're moving quicker than they were before. And we're seeing that, that investment, those kinds of, you know, dedicated teams are all pushing towards the same goal of embodied artificial general intelligence, which will be very, very fascinating to see how society manages to digest that one. Now, this is the video I definitely want to show you guys, because this device, the humane AI pin, which got a lot of hate recently, and I say recently, it was probably a couple of months ago. But the point is, is that they've made a new list of updates that I think people are going to want to see.

Hi, everyone. Greetings from New York. We've got a few updates to share with you today. First, AI pins are scheduled to start leaving the factory at the end of March. Priority access orders will begin delivering in mid April and the rest will follow shortly after. We'll be sending onboarding emails before pins go out, and we'll let you know when to look out for those next. Everyone that has purchased an AI pin is getting the first three months of your humane subscription on us. It's a token of Gratitude. It's our way of saying thanks to our early adopters. And the offer extends until March 31st. So buy an AI pin today and get the first three months of your Humane subscription for free. Your Humane subscription includes over the era software updates, meaning your AI PIN gets better over time. We'll share more about our planned feature releases soon. We've also enabled free shipping and if you bought a PIN already, don't worry, we'll take care of you next. We've loved the response we seen to the content that the team's been putting out recently and Colin and I are here in New York to show more friends AI pin. So if you want to see this thing in action, drop us a line. And lastly, shout out to our discord. Thanks for being on this journey with us.

So let me know what you guys think about this AI pin. Do you guys think this is going to catch on? Is it just a wearable? Is it just a fad piece of technology just going to be obsolete within the next 10 to 20 years? I genuinely have no idea. But I do think that AI wearable tech is here to stay. And an interesting point that some people have made is that why can't you just put that technology into a phone? And I think it will be interesting to see the first company that does try to do this. But I also do think that wearables are becoming the new, I guess you could say, piece of technology that is cooler to wear because ever since the Apple Vision Pro, people are becoming more and more accustomed to seeing people wear technology that is out in the open. So I will be paying attention to this kind of technology. I'm not sure if I do want to buy this just yet. I kind of do want to see some first reviews because it is the first iteration of a product and that product, the first iteration, is always improved in the second generation. So I usually just wait for the second generation if I don't absolutely need this. So it will be interesting to see how this product is developed and how it does change over time. And also an interesting clip from Eric Schmidt saying that this is the future. He's saying, text to action. Take a look at this.

Now the way it will work is it'll be basically text to action. You'll have an idea and you'll say, I want a this. And the system will show you the recipe or organize the events and the systems will be smart enough to be able to communicate, send emails, make phone calls and so forth. Now, the way it will work is it'll be Basically text to action. You'll have an idea and you'll say, I want a this. And the system will show you the recipe or organize the event. And the systems will be smart enough to be able to communicate, send emails, make phone calls and so forth. Now the way it will work.

So let me know what you guys think about that. I think we're definitely moving into a future where it's going to become harder and harder to predict things. I mean, text a video was something that I genuinely believed was three years away, at least based on the current technology. But it's clear to show us that the, you know, exponential returns. Trying to plan for things is just very, very hard to do in the field of AI. And we can only look at those who've done it accurately with respect. So I would say text action could definitely happen. Anything's possible in the realm of AI, but this is definitely something that's quite hard to conceptualize considering we don't have agents yet. But I think we could definitely get there very soon based on the amount of things that are going on behind closed doors with AI agents. So although this kind of tech does seem pretty crazy, it is something that will be arriving sooner than you think. We also had Xiaomi do AI portraits essentially. They released a new phone and they had this presentation in which I believe they're essentially making, I guess you could say like a kind of tray grained image of a person and then making like these perfect images of the person. It's actually really cool model of the subject. This is an AI portrait. It recreates a photorealistic likeness of them.

That the AI can then work with.

With this AI portrait, you can now.

Submit a prompt for the software to generate a brand new, totally unique image.

Let me know if you think this is dystopian or if you think that this is just a really cool piece of tech that you would like to have and you'd like to use. I think this is something that's, you know, pretty harmless. I think it's. As long as it's used in, you know, you know, safe ways. I think this tech is going to become really fun for people who just want to kind of, you know, be creative, kind of want to see themselves in different locations. Maybe you can't travel there. Then again, this phone is pretty expensive, so you could just book a flight. But the point is this tech. I do think applications like this are really kind of fun. Then we had the CEO of Nvidia state something that was very, very controversial. This got 7.4 million views. And it was quite controversial because he said something that challenges the belief of many individuals way of living. So he basically said that people should stop learning how to code. Like if you're, you know, a kid and you're going into coding, it's kind of pointless. And you know, I'm going to show you guys the point first and then I'm going to give you guys my rebuttal.

Something it's, it's going to sound completely opposite of what people feel. Over the course of the last 10 years, 15 years, almost everybody who sits on the stage like this would tell you it is vital that your children learn computer science. Everybody should learn how to program. And in fact, it's almost exactly the opposite. It is our job to create computing technology such that nobody has to program and that the programming language is human. Everybody in the world is now a programmer. This is the miracle of artificial intelligence. The countries, the people that understand how to solve a domain problem in digital biology, or in education of young people, or in manufacturing or in farming, those people who understand domain expertise now can utilize technology that is readily available to you. You now have a computer that will do what you tell it to do. It is vital that we upskill everyone. And the upskilling process I, I believe will be delightful, surprising.

So I mean, this was rather controversial because a lot of people are saying that, you know, just know, English is not a sound programming language. And then some people saying kids should stop learning to write because we have a printing press, some medieval dude. I mean, I think we're at a very confusing time because if you try and look at the next 50 to 25 years, I mean, it seems like everyone's job's going away anyways. So I mean, you know, this point can be argued and I guess it does make sense if they're able to do, you know, human level code with these AI systems, especially if you've got like an AGI system that's able to completely just bust out some code. And we know that jobs that are on a computer are some of the most vulnerable. So I mean, it is a very, very hard thing to do. But we do also know there was like a recent research study that showed that the recent code developed by LLMs was less secure. I mean, I think this is going to be something that time will tell if this is true or not. I do believe that in the future there's still going to be coders, still going to be software developers that are going to develop new frameworks, new languages, new things to kind of, you know, work with. So I do think people will still be around for quite some time. And I do think, whilst we are moving towards a paradigm where natural language is going to be a form of communication, rather than me, you know, walking around, checking my notifications, you know, checking the DMs, you know, checking YouTube notifications, uploading a video, rather than just saying, upload a video about this or that. I mean, I think until we get to a very comprehensive AGI system, I think people should still definitely pursue those careers, because I think as they evolve, people will be able to evolve, too. And I think it will be interesting to see how that does happen. But I do want to see your point on this, because this is definitely a very controversial.