AI models function as massive spreadsheets containing billions or trillions of parameters (numbers) that process information through next-token prediction, where the model predicts the most likely next word in a sequence. While AI can generate sophisticated responses, it fundamentally operates by completing patterns rather than true thinking, and after initial training, it stops learning and relies on external memory systems. The technology presents both significant opportunities as a great equalizer providing access to expertise like doctors and lawyers for everyone, and substantial risks including mass surveillance capabilities, potential job displacement in white-collar sectors, and the challenge of balancing benefits against ethical concerns. The key insight is that AI, like all transformative technologies (fire, printing press, internet), exists on a spectrum of good and bad rather than being purely utopian or dystopian.
Inmersión profunda
Prerrequisito
- No hay datos disponibles.
Próximos pasos
- No hay datos disponibles.
Inmersión profunda
Is AI Evil? Are Humans Aliens? We Went Deep | Pilot Episode | NocturnalAñadido:
It's showtime. Hey, welcome to the first episode of Nocturnal here at the New American Late Night. What we're going to do here, we've never done this before and we're very excited about it. On my left and right are the two legends. So, let's start with the one who's never been on camera before. Of course, it's Justin over there. Look at him.
>> We're so goddamn excited. And of course, you guys know Sulie 4Q over there.
Legendary serial entrepreneur, artist, producer, real estate owner. He does too many things. I can't even list them all.
But we're here in this Nocturnal to do Nocturnal stuff. It is literally, guys, what time is it? Looking up the clock right there. It is 1:15 a.m. here locally. This is when you record Nocturnal.
>> This is what Nocturnal is all about.
>> So, let's dive straight into it. AI Justin AI because this guy right to my right, but he's an AI wizard in my opinion. Like Harry Potter classic wizard. He knows so much about AI. You guys were talking about so many great things back, you know, before we started recording. What is it? What do What's festering on your mind with AI, though?
Go ahead. There there's just so much to say about it. I think >> when you asked what is it, I think that's actually a good opening point here.
>> Okay.
>> Because what AI actually is, >> it's a bunch of [ __ ] numbers.
>> Yeah.
>> Ones and zeros.
>> No, it's not ones and zeros. Think of think of a Google spreadsheet.
>> Okay.
>> Like think of like Excel. Microsoft Excel. Everyone knows Microsoft Excel.
Think like that.
And in all the little uh cells you just have numbers and these numbers look like 0103567.
>> Um >> is that pi?
>> That is not pi. A little bit off pie.
And they're all different numbers, right? And these are um these are parameters or weights in a model. And so just imagine billion of these things.
Sometimes >> billion cells >> billion cells of these numbers.
Sometimes trillions. The the latest models have trillions. Okay. And I'm trying to simplify this as much as possible, but >> doing a great job.
>> The idea is you put, think of it conceptually. You have this this Excel spreadsheet with a billion numbers in it.
>> And somehow you feed it a question.
>> Mhm.
>> Um I'm oversimplifying this. It's not exactly how it works, but you feed it a question. It goes through this [ __ ] Excel sheet of a billion numbers.
>> Yeah.
>> And then it outputs an answer intelligence, which if you think about it, our our AI overlords is a spreadsheet of a billion numbers. Like that's that's trippy.
>> Yeah. Well, I've heard it described like this and you tell me if this is accurate or not that it's basically a a immaculate hyped up make makeup on version of completing the sentence of of it's basically just it's finishing a sentence. Yes.
>> Textually. It's not like actually thinking so much as it's it's predicting the next word in a sentence. Is that is that accurate?
>> That's exactly right. There's lots of debate in the industry around that.
That's how it started. It's called next token prediction.
>> And that's how they train these models.
So if you say um um we went to the farm and we were thirsty for milk, so we milked the >> it's going to say cow.
>> It's going to say cow and it just >> and >> second the second or second like most likely answer would be goat or something, right?
>> Exactly. And so that's how it started with these earlier models. But what they found is that as they start doing more and more of this, it's actually not just next token prediction cuz in order to predict the next word, >> yeah, >> you need to have a semantic understanding. You need to have this deep understanding of the world and a world model and all this. So >> without getting too esoteric like >> um these models yes they are predicting the next world but there is some weird intelligence beneath it all that to be truthful >> uh most of us in in the industry and and all of humanity don't understand exactly what's happening here and this is >> cuz it's constantly learning as well. So it's fed an initial knowledge base. Am I correct? And then it's learning as us as humans continue to input into it and then it's growing into something that we don't fully understand yet. Th >> this is a very nuanced topic and it's a very um timely question um so the way they work is >> and again this is coming from a guy who who headed up Google's innovation of AI by the way. I I was a product lead at Google um for AI innovation.
>> He know B he knows what he's talking about essentially.
>> It was just a little background on that.
So it was this was right at the forefront of when AI came out chat GBT and all that and Google put a special task team together orchestrated by the executives and by the CEO. Um and the team I was on um was called AI product expansion. And what we did is we took the latest and greatest from AI. We worked with our research scientists at Google, some of the smartest people in the world.
>> Um, we saw what was coming, what they're working on, and then we had to take these AI applications and bring them into YouTube or Google search or Google Chrome or Gemini um or Gmail, all these different products. So, that's what I did. Um, but anyways, just to go back to Sully Singh real quick and then I'll I'll stop yapping and you guys can go on. But, um, >> you're fine. Go.
>> So, the way it works is you give these models a [ __ ] ton of knowledge. the internet's text, all this stuff, >> right?
>> And it it learns that >> trains it. But here's the catch.
>> After it learns it once, they train the model, >> it stops learning completely at that point. Think of it just like a cut off point.
>> And so basically, anytime you ask it a question, it just it's stuck in that knowledge that you trained it on already.
>> So, all right. It it sounds like as soon as you stop training, it stops learning.
>> Okay.
Well, obviously go ahead.
>> Uh, so to simplify for some of our audience that aren't, you know, in the field that we're in. So then after it stops learning, now it's based off memory.
Am I right?
>> Yes. So, >> so, so it's taken and it's learning more and more about the users and then it's taking that gained memory and utilizing that to better answer questions.
>> That's right. slowly but think of that as outside the AI brain. Think of that as like a little u memory bank almost.
Think of it like a hard drive that the AI model can just hit and say give me some information for that. So it's not actually in the AI brain those memories.
It's external and they can fetch it.
>> But but I want to say one thing here because Sulie's question was interesting. That's where the industry was at for a long time.
>> Mhm.
>> You train a model, it learns, it's got everything in it and then people use it.
Then you train the next model bigger learns more.
>> The big thing now is called continuous learning.
>> Because if you know human beings and our brains, >> we don't just like come out the womb with all the knowledge and never learn again. We're continually learning throughout our lifetimes.
>> You know, you read a book or you see the matrix, they download stuff in the brain. You do that by reading books, by having conversations with people. AI models cannot do that.
>> So, so that's the next big breakthrough that all the labs are working on.
>> Well, let me jump in here. My question is uh I think may touch on what Sulie was was talking about when we talk about what these these models learn from user input. So they they get trained by the companies but they're also am I right or wrong also being trained by the data from the customers. So every time that I use the product it's also improving itself. Right. So is that a separate brain? Is that kind of what you were saying?
>> No no that's right. The key distinction is like when this is happening.
>> Okay. So as you're chatting with the model, >> yeah, >> it's not learning like it doesn't update its brain, right?
>> It's not real time.
>> But what they do is they take that data.
>> They take that data that says solely thumbs up this one or Matt >> anonymous user.
>> And then the next time they do a training run, >> okay, what is a training run?
>> Training runs when they create these brains.
>> It's creating the next model.
>> The next one. So the next model, you know, from going from >> so that's when they so they use all the user input. They bring that in as new data and learns. But it's not like when you thumbs up the AI, it updates its weight and all a sudden learns. But that's where websockets industry wants to go and that's what they're working on.
>> Um, >> close. I would imagine pretty close.
>> I don't know if close is the right word, but I'm hearing lots of whispers of confidence that um >> that it's scary when you say confidence with AI. So, for some reason that freaks me out.
>> Well, that that models will do that that they will eventually learn that. So well uh here's my question because you know AI has taken over the world. We all know that that's not news and it's it's industry after industry. It's you know started with graphic designers and so forth and now it's you know doctors, lawyers, etc. What is it? Is there anything to you though Justin as someone so knowledgeable and like you're so deep into this? Is there anything that to you that even scares you? Cuz I feel like and tell me if I'm wrong. very optimistic about this and I am too.
>> I think Sulie is probably pretty optimistic about this as well.
>> But even to you though, you know, is there anything that you're not optimistic about?
>> I I will have to go on a rant to answer this question.
>> Proceed.
>> Rant on.
>> It is nocturnal.
>> Um I have so many thoughts on this and this is kind of the crux of the conversation and they I love them >> and myself I am very optimistic, very excited.
>> Get her done. I think that this has this will bring bring great benefit to society. So I >> which is in contrast to a lot of the opinions out there.
>> Well, here it's nuance though.
>> In typical form it's [ __ ] black and white right now. It's it's Republican, Democrat, Yankees, uh, Red Sox. It's like >> everyone's all on one side of the other.
It's like no, it's not AI's all good and oo utopia or AI's all bad and dystopian.
It's like there's going to be good and bad. And I don't know why gray areas are so hard for people to understand. But if you look at the fabric of reality, the whole concept of ying and yang is that in the fabric of reality there is good and evil. It's so deep. It's like deep.
>> There's a spectrum. That's what you're saying. There's a spectrum.
>> There's a spectrum and they they they go with each other. You can't have one or the other. And so so here's some examples.
>> It's like the matrix. And I promise I'll I'll bring this all together. But the first technology humanity ever invented is what?
>> Fire.
>> Right answer.
>> Nailed it.
>> Right answer.
>> Think Think of how do I win? By the way, think of how magical that was. Imagine like you're a caveman. All a sudden someone figured out how to just harness fire. They would be considered magicians. Like whoa, that dude just created fire. But so here's the thing.
>> So here's >> I like where this is going.
>> This is the first technology. Fire could keep them warm at night. It could give them light at night, >> kill bacteria on food.
>> Yes, it can do all sorts. But it can also burn down villages.
>> It can it can create chaos and create disaster. It can kill >> scary.
>> You have you have the yin-yang. You have the good and bad. Okay. So then let's just kind of continue down this this progression. So we had fire and then >> fast forward way further. You have printing press invented books. Okay, great. Spreading knowledge everywhere.
But it's like, okay, >> people are reading about Mine Comp and people are reading about these like questionable books or, you know, like is that a good thing or a bad thing? So, like pros and cons. You get to radio, you get to TV, you get to the internet.
Same [ __ ] You get to social media. You get to connect with all these people on a daily basis. You can see what your high school mate had for lunch yesterday.
>> Um, but then there's addiction stuff.
There's all sorts of messes and issues with social media. AI will be no [ __ ] different. There will be good to it and there will be bad to it.
>> I agree with this wholeheartedly.
>> And I believe that the uses of AI just same as fire, you know, imagine not being able to cook. You know what I'm saying? The uses of it are so useful that you will tolerate what comes with it. Now, as humans, we love to simplify our lives. everything that's, you know, been invented to simplify whatever we have to do on a day-to-day basis has, you know, transformed, you know, today.
So, I feel like with AI, it will simplify our day-to-day task, so on and so forth so much to where we we're benefit from it so much that you almost have to take what's come what comes with it. And then what happens is new inventions and things happen that help mitigate >> the things that come with it because, you know, there's fire safety now. Yep.
>> Well, you know, Sulie, you you gave me I I believe it was you a great example >> uh last time we talked about this.
>> Uh you know, I'm graphic designer kind of by nature and and you brought up what I think is the best example ever when Photoshop came out. That was you telling me that, right? Why don't you go ahead take it from here?
>> Yeah. Cuz you know, as a artist, so I'm a visual artist. I used to draw, paint, things of that nature. don't do it as much now, but when I was younger, I was in an art high school. And at the school, Photoshop was just becoming prominent. So, as us visual artists, we're all painting and things of that nature. Some of us started doing our paintings and drawing digitally. And the the pure artists, our fellow classmates were really upset about it. They felt like it was cheating.
>> Cheating. Yeah.
>> You know what I'm saying? And and it's the same way people feel in music and stuff now, too.
>> It's how they feel about AI. They feel that way with AI. Like you got to write a book. Someone uses AI is cheating.
>> But the winners apparently >> But everyone on that adopted the new technology.
>> Whoever adopted Photoshop early had unique opportunities that the rest of the artists did not. And those pure artists, a lot of them got left behind.
>> Yeah. Absolutely.
>> Because they never learned the tools of the now. And hey, look, we were just [ __ ] around in Photoshop 15 minutes ago, literally.
>> You know, it's our first episode of Nocturnal, but you know, we were [ __ ] around with our with our logo and stuff.
And you know there's AI built into there and it's just it's a great kind of circling back to all of this like I really do think and that that one hit home so hard with me as a graphic designer cuz the graphic designers were like some of the first people to get [ __ ] by AI >> and you can see it happening right now like everyone who's an early adopter of AI and found you know creative ways to integrate it into their business you see them winning big right now and those who are being resistance to change. Those are they're going to be blockbusters.
You know, >> wasn't it Justin? Was it you that told me uh I think it was you that you know the the best people with AI are the people that were designers or something.
>> I think that I I think creatives are going to absolutely [ __ ] thrive in this environment.
>> I mean, why is that though? Cuz there's so many graphic designers, video, you know, whatever. Because you creatives have a toolbox available to them for the first time in history where they can express their creative visions, genius, taste, like almost pixel perfect from what they see in their head to what comes out when they don't have to hire people.
>> Don't have to hire people. You don't need all that money. You can do things in your kind of through your own mind's eye. And like >> creatives are going to thrive. Let me take this even um a step further to be more specific. people with good taste is what it's going to come down to because you can have an AI create any website, any song, any image.
>> Yeah.
>> But like you see >> and and it was back off what you said.
It was the same thing with Photoshop because if you were already a great visual artist >> and then you went to start using Photoshop, the things that you created was far different than someone who didn't have good taste. Y taste now with the mind, >> you're going to win big.
>> Well, here's something and I have have talked about this over the years. I wanted to bring this up to you, Sulie, >> which is Justin and I have had a lot of conversation over the years of good taste >> like me, which he just brought up and I know that you have this which is why I wanted to bring it up. I don't know the best way to like I don't know even communicate to him Justin though like cuz cuz you were the one that even made me aware of like the the whole good taste phenomenon with creators like how would you describe it to him in the same way you know cuz you kept bringing up to me like this is what good taste means.
>> I just it's it's one of the most important things in being a creative >> right? Um, I mean, look, you >> it's it's not easy. I'll put you on the spot, but you like, you know, >> I want to tell a story.
>> Mhm.
>> And I think this this will come across critical, but that's okay. I I went to a job interview one time with NFL.com.
>> Get her in.
>> And at this point, I was kind of >> NFL.com.
>> Yeah. NFL.com.
>> He doesn't even watch the NFL. It's crazy. He really does. and he watches the NBA.
>> At this point, I was junior in my career, so I couldn't speak like esoteric language, but I had created a number one app in the world that went to number one on the iPhone, 7 million downloads. I I'd done all these things were like really successful. I went and I I went for this job interview at nfl.com and it was this director there.
This sounds [ __ ] up and mean, but I'm just being honest with you guys. The guy like he dressed like a [ __ ] turkey. like he just like damn >> it it looked it just it was like you know the old guy wearing like cargo shorts still it's like it was just so bad >> affliction t-shirt >> and I'm just saying like this guy has such poor taste and it doesn't all you know maybe different domains you can have different taste but for the most part it goes across I'm like there's no way he has good taste with his digital endeavor stuff so the whole reason I'm going on this tangent is like ta >> t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t taste is extremely important for the creative cuz anyone can create, anyone can have ideas, especially in this AI world, but the ones that are going to win are ones that people have good taste with, make good decisions with, know what people like.
>> I think it's really difficult to put your thumb on what that means. And that is hard.
>> I think to simplify it, I believe that good taste is something that's um generally accepted. So, so for instance, you know, someone puts on a nice suit.
If majority people agree that this is a nice suit, that's good taste.
>> Yeah.
>> You get what I'm saying? So, if the guy that he's describing was badly dressed and majority of the people that see him would agree that he is not dressed, >> he doesn't have good taste.
>> He doesn't have good taste. So, I think that's the best way to >> Well, when we were when we were in the booth here >> and we're all collectively putting our minds together, just kind of [ __ ] around with, hey, an AI logo, >> that that was in in my opinion the epitome of like this is how you know people have good taste because >> I I love being around creatives.
>> And I know you guys do too. it. But like I think we all know there are people that when they throw out ideas, you listen and there are people like you respectfully listen and you just kind of like thank you for that.
>> Yeah, I appreciate >> like you guys are always people that you're like, "Yeah, okay. I'm actually listening."
>> Like what? Okay. Like I'm I'm tuning in.
My frequency is up. My ears are perked up. I'm actually like cognizant and tuned in to this. But they're, you know, you're good friends, you know, very good friends. Julie, you're in so many creative things. I'm sure you have tons of friends giving you feedback on music or whatever you're doing. And there probably some people you're like, "These are people I listen to." And it's actually rare like, >> you know, you get if you are someone that most people would say has good taste. Uh it's you're normally the guy that people are coming to and asking, you know, your opinion of things.
>> Absolutely. That's so true. you know, but the opinions that you're getting aren't opinions that you're asking for.
>> You what I'm saying? So, you know, >> interesting. That's very interesting.
>> You know, so that's, you know, that's the best way to put it. It's like people are asking if you have good taste, people tend to ask you for your opinion.
And people who tend to not have good taste, they offer their opinion un unprovoked. You know, >> 100%. Hey, hey, I I don't want to sidetrack this, but did you know, Justin, that this [ __ ] over here, girls had his picture in their lockers in school?
>> You going to put that on camera?
>> Damn. Tell me more, man.
>> Well, you know, I'm a happily married man.
>> This is a long time ago. They literally did. There were literally pictures of him and girls lock. Can you imagine that?
>> I'm not surprised. Like I said, like I said, man, you know, it it some people looked at me in a certain way, but then, you know, I I might ruin it anyway, so who cares, you know?
>> Real one, man.
>> I know. I know. It's so humble and genuine, too. I can't imagine it.
>> Can you Justin, if there were girls in high school had pictures of you in their life to an all boys high school, so that was >> Get her done.
>> That was not a thing.
>> Yeah, there were a few boys their picture. I guarantee you.
Oh man.
>> But man, so what do y'all think uh the world will look like 10 years from now with AI? Like what jobs are out of here?
>> Well, is it strictly AI or is it the combination of AI with cybernetics and you know Optimus robot and everything else, >> technology in general.
>> I'm I'm going to get this just >> I have lots of thoughts. So this this actually continues the conversation we had earlier pretty well because >> there's going to be lots of good that comes out of this and some bad, too. And I I think I have a decent um vision of what both of those could look like. So, let's start with the good stuff. I think AI is going to be the great equalizer.
>> A big hot topic in our society right now is economic disparity, all sorts of disparity across kind of um the societal um classes and scales. I think AI is the ultimate equalizer. I think when everyone has in their [ __ ] pocket in their phone, everyone in the United States, everyone in low class, middle class, high class, everyone in Africa, everyone in different countries has access to the best doctor in the entire world, meaning AI, >> the best lawyer in the entire world, >> which is AI >> through AI. the best software developer in the entire world that can build out their dreams, accountant, finance, tax, psychologist, >> which is AI, >> all of these things that to me is the great equalizer and I think that's a beautiful thing in this world and I I think we got a taste of that with the internet. I think we got a little taste of that with the internet.
>> So big question, what happens to college?
>> Oh, that's a great question.
>> Oh my god. I think I I think college is already obsolete with the one exception that it's a great social experience.
It's a phenomenal experience to meet friends, >> to socialize with the opposite sex, >> especially if you come from all boys high schools.
>> Yeah.
>> Like both of us did.
>> And it's like I'm sure all of you guys, all of us in here, every single person in this room can relate to, we still have friends that we met in college.
Probably deep friendships.
>> Some of the best. But I don't know if you guys are anything like me like I learned a little bit but >> nothing really >> knowledge wise very little came from that experience is the people that you meet >> exactly and today I think I swear by this and I wish I known would have known this a long time ago but you can go to YouTube University right now and learn anything you want >> anything you want man and that you follow your interest you want to learn physics you want to learn about how to um uh restore an old car or a car engine. Uh how to do your plumbing, anything you can >> how to build a studio like I did here.
Yeah. Yeah. Literally.
>> So I I do think it's obsolete solely. I think um the masses are still sending their kids to college, which is fine. I think in 10 years that probably looks pretty different. I think certainly every kid will have their own AI tutor >> that goes to their own progress.
And for those parents, >> that one hit hard with me >> right now.
>> Some of the most wasted time of your life a lot of times is those first 12 years where you're just stuck in those classrooms >> learning all the things that you're never going to use in your life, you know, not just college, but yeah, cursive is gone.
>> You know what I'm saying? Like I thought I was going to use this forever.
>> If you, you know, if you become a influencer or famous and someone asks for your signature, it's helpful. You might have a pretty signature, but what else did you need curses for? Or or a lot of the topics that you learned in high school outside of, you know, math >> and, you know, some some of the basic things that did not take you that long to learn, >> you know, what does that look like now?
Like my kids are homeschooled, you know, with that same thought.
>> That's the move solely. I feel like that's the future.
>> Yeah.
>> And and your wife teaches them, right?
>> Yeah. And and but but you know what's crazy is if you really think about it, >> the the best tool that you could learn as a kid is how to teach yourself with the tools that are available to you.
>> So true, man. So true.
>> You know, you you need to be able to ask questions >> to learn on your own.
>> To learn on your own >> because that's what creates continuous learning.
>> Well, can I jump in here and ask you guys, >> do do you feel like the best creators are selftaught? Cuz I do.
>> Yes.
>> Yeah, >> I do too. And I think that's what both of you guys are.
>> It takes a it takes a certain uh thought process to be first being a selfarter is valuable in every industry.
>> Yes.
>> You know, that's across the board.
That's a rare trait in itself.
>> Yeah. A problem is not an an obstacle you can't overcome. The problem is, okay, I got this. I'll figure this out.
>> It's self-evident though, man. Like if if you choose something you're interested in, you choose to learn it and you go through that process, you're just going to learn so much deeper, so much more. It's just it's just very obvious that that that is the best.
We're also biased because all three of us kind of are selftaught, you know, but >> yeah, I do think that is the route.
>> Well, I mean, when I think you know the people that Yeah, I think there is a bias there because I'm sure there are some people that aren't self-taught that's that are still self-taught uh starters and have done very well for themselves.
>> It's not all or nothing.
>> Yeah. you know, but I do think some of the best creators that I've ever met >> have been selftaught >> across the board. Big world out there, so you know.
>> True, true.
>> Yeah.
>> Well, let's take Hey, ladies and gentlemen, of course, the first episode of Nocturnal. We're going to take a quick break. We'll be right back after this. Don't go anywhere.
Hey, welcome back to Nocturnal, the first episode. We're so glad you guys are here. for keeping the vibes cool and smooth like Justin's lack of beard over there. Look at it over there. Of course, Sulie 4Q, who we failed to mention how much of an AI legend he also is. Sulie, can you tell us about, you know, your your software engineering and your experience and knowledge there?
>> Okay. So, you know, take it back to 2021. Don't want to bring the mood down, but almost died. Had blood clots in my lungs, lost my ability to walk and use my voice. I'm a rapper by trade, so losing your voice is an ideal for hitting the stage or recording music >> or doing behind the scenes work, sending off reference work or any of those things. So, what do you do? Do you just curl up in the in the corner and cry yourself to sleep and wait to die, or do you think of how what can I do? So, I could still get on my computer. So, I'm completely self-taught, full stack developer. I already had a design background at that time. I dove deep into learning, you know, the back end of the development side and learned a ton about AI as it was becoming more prominent and, you know, it definitely helped me shift gears until I could get my voice back and uh start walking again and kept me positive. Law of attraction is real, people. That's just my opinion.
>> Hey, that's incredible. And it also ties back into what we talked about earlier, which is the self-taught stuff.
>> Yeah. Yep. Right on. And you you you turned a negative into a positive as well.
>> Yeah, man. And and I'm a law forward, man. So, I feel like, you know, no matter what you're going through out there, you know, uh with a positive mental attitude and a work ethic and and a how can I attitude, >> you know, you'd be surprised what you'd be able to do, man.
>> Thanks.
>> That's incredible. Appreciate it.
>> It really is. He went from, you know, you know, puking up blood, couldn't walk.
>> Yeah. to a unexpected blindsided scenario that they he then powered through went through a year or more of what was it daily therapy for all kinds of things.
>> Well, well, you know, uh more just heavily medicated.
>> Yeah. Uh definitely went into rocky mode. So I kind of >> crawled around the house and picked myself up to >> Yeah.
>> to use the restroom. You >> sure you know >> that's real though. Uh definitely a lot of open water bottle attempts, failed attempts that I had to painfully ask my wife to help me open the water bottles until, you know, I finally could get those open and called my boy Chris Calico when I got my first push up in.
>> Hell yeah.
>> You know what I'm saying? And I remember hitting that first lap around the backyard, man. And I hit that lap around the backyard. I knew it was on from there.
>> I said, you know what?
>> Yeah. Wings up. I'm done with this.
>> That's all.
>> That's incredible.
>> I'm back.
>> Yeah.
>> Start throwing punches in the air, you know. I'm back. I can't be stopped, >> you know.
Okay. It is nocturnal though. We got to throw it down a little bit. Justin though we didn't want to, you know, you mentioned something off air here a minute ago that you did not want to leave the AI conversation hanging dead here without you being able to communicate a few things. There are there may be some negatives here. Go ahead.
>> Yeah, I mean there there needs to be a balance to this. I am >> a balance in the force like Star Wars.
>> And I I am quite optimistic. I I do think a lot of good things are going to come. I think when we potentially cure cancer or solve certain diseases, I I I try to not be too spiteful when I say this, but a lot of the people knocking on AI right now and crapping on AI should uh >> you can say [ __ ] >> [ __ ] on AI like they I mean they should be careful because when these things end up having huge benefits for humanity, they're going to be the first in line >> to take advantage of that. So >> anyways, but there there are downsides to this as we discussed. There certainly are. This is a gray area. It's not black and white.
>> And after kind of lots of thinking, I see a few different downsides. I see three to be precise.
>> Um I think I think the biggest one the whole concept of big brother gets put on mega steroids and mega peptides. Now, >> for example, in China, they have 700 to 800 million closed caption cameras across their society. When you load those cameras up with some really simple, lightweight AI software that does facial detection and can do all the sentiment analysis and all this different analysis. And you know, at that point, you know where everyone's at at all times, what they're doing, what they're not doing.
>> And um >> that's that's a bad thing. Think of think of your tax returns. Historically, 1% or half a percent of people got uh audited by the IRS in their tax returns.
>> Well, that number is going to go up to 100% now with AI because rather than humans having to audit this, you just feed it to an AI system for sense and now 100%. So big brother and the government is going to have these superpowers just as the individuals are. I think that's a a huge downside of the AI scene. Can I jump in here though because my biggest thought and we've talked about this on our late night show is it is my fear and I think this is a pretty common thread is not the United States even if the United States does everything perfect. My fear is what the other countries do. So even if we control all of our AI companies perfectly, we put all the procedures, the rules, the safeguards in place, >> what about China? What about Russia?
What about India, >> Australia? Like Argent Argent, who knows? It doesn't there's even if we do it perfectly, someone else is just statist statistically going to [ __ ] this up.
>> I I totally agree. Totally agree. I I think this is all moot point. I think the genie's out of the [ __ ] bottle.
It's going >> like it's gonna unfold anyway. But just let me just let me these two things real quick. So >> big brother is a big one. Second is this drone technology. I do find that to be kind of terrifying. But when people think of big bad AI, they think of Terminator and all this and that and like robots coming to kill us.
>> Imagine waking up and there's a drone at your window.
>> Yes. And it looks like Noah >> and it like speeds off after you wake up, but you just seen a drone just staring at you, sleeping.
>> This could be the version of Terminator.
Instead of robots, they're drones. And >> you're already seeing this in warfare with Ukraine and Russia and um >> Oh, yeah.
>> Certainly the future of warfare, but >> Well, it is though.
>> Just these swarming drones with AI brains. And and here's the big catch. If you're driving, this is a thought experiment. If you're if you're I love the points, man.
>> Well, I'm pointing cuz when I switch cameras, he doesn't All right. I'm pointing to let people know they're on camera.
>> Okay. Got it.
>> Stop pointing because your reflection's going off on his screen.
>> All right. All right. Well, looks great.
>> Spielberg over here.
>> So, when you're This is a thought experiment. When you're driving in the car, you have your pregnant wife with you. M >> she's about to go into labor and um you know you're you're going to the hospital, you're driving 80 mph in a 35 mph zone. Today you get pulled over, cop pulls you over and you're like, "Dude, my wife is going to labor right now."
And he's like, "Oh, I totally understand. Like, let's get you there, bro. Escort you."
>> If you have these AI cops or drones or robots, you you you can't and and they're programmed to you cannot speed over 80 miles per hour or kill if you sort of see where this is going.
>> You can't negotiate with that.
However, there's no negotiation.
>> Wait, hold on. However, do you don't think by the time that we get to this this AI, >> not only do are they operating vehicles?
You know, there's apparently some type of humanoid robot that's going to get out of there, check your ID or whatever.
>> Yeah.
>> You really don't think that that point the training there does involve the human factor?
>> No. Like, who's training them? Is it is it New York Police Department that's training them? Who's training them?
technology wise, that's so difficult to implement because humans could just lie and and and how would they be good lie detectors and and you know they could assume that you're lying when you're not lying or vice versa. So that would be a slippery slope and you also don't want to give them the ability >> to decide on their own because they decide on their own, okay, let me escort this person to the hospital. They're good. But then the next person, oh, they're armed and dangerous. Let me shoot them. You know what I'm saying?
Even though that's against their train, that would be against their exact cops [ __ ] this up today, too.
>> Wait, but this is going exactly where my next question was. Do you really think that these AI controlled humanoid robots with armed lethal weapons is right around the corner? No.
>> No, it's not.
>> They're not going to have lethal They're not going to be out there with real loaded guns. I don't think so.
>> I think you're right. The drone piece is interesting. Just imagine, imagine this.
>> Imagine this very simple uh drone with very simple AI technology that has, you know, find this double O >> find Ollie Olsen face and kill. That's the two simple commands. Stop it.
Nothing else. And then it just goes and it it it just that's its entire existence, you know, and that's terrifying.
>> Well, and here's another >> that would definitely probably be on the black market for sure.
>> Yeah. Yeah.
>> You know what I mean? H.
>> But uh Justin, this is a famous example I heard. I'm going to butcher it. And Sulie, I'm assuming you probably heard this as well.
>> It's like I think one of the biggest fears though is you know if we train AI to the point of uh let's say there's one singular goal and that goal is make as many paper clips as possible, right? And so in order to do that, well, if we could make more paper clips if we cut off the water supply to the humans and we could make more paper clips if we turn off the uh nuclear power generators for the cities cuz that would allow the AI to make more paper clips. So if you give it a bad prompt essentially, but in my mind it it is a great example. It's very thoughtprovoking, >> but I do feel like it's it's very I don't know lowhanging fruit.
>> I feel like the Debbie Downers are extremely valuable right now.
>> Okay.
>> All your pessimistic thinkers, you need to have them in the room at every you know development meeting cuz they're going to think of all the worst and terrible things that could possibly happen. And then you also have your problem solvers in the room and then they take that list. I think >> these these people are problem solvers in >> love. The nocturnal is problem solvers.
>> I think that's really insightful. And the last thing, Ollie, I think I think you're right. Um the first two are a little, you know, maybe sci-fi out there, but there's one right now, this is the last big downside is this whole idea of job loss.
>> Job of the >> job loss.
>> Job loss. Okay.
>> Up up up until about six months ago, I thought that was all overblown and [ __ ] I was like, we'll find new jobs. But I am seeing it so clear as day right now that white collar is about to get [ __ ] obliterated.
>> And I think that is a real tangible outcome negative side effect of AI that's going to happen in the relatively near-term future. I think it's happening right now and it's going to progress and escalate and that brings up a whole sort of question.
>> It's escalating so fast. It's escalating so fast >> that I don't know if we're problem solving fast enough in order to help prevent that from happening.
>> Strong point.
>> So, so it you may start seeing, you know, people really suffer from job loss before you see the solution.
>> Yes, absolutely. And Justin, as someone that that was deep in that corporate environment with AI, do you think there's any chance that the recuperation of jobs, people being hired by AI companies can compensate with the job loss or is it going to be so massive the delineation here? like all the people losing jobs to AI. Are there enough people being hired by AI companies to make up for it or is it just so you know? Go ahead.
>> I think the answer is definitely no to that. I think we're going to see a huge trend of very small companies, single person companies, two or three people companies doing tens of millions, hundreds of millions of dollars. And I think um yeah, I think there's going to be a massive job shortage. and and I don't >> job. Wow.
>> And I don't think that's a good thing.
>> Um I think people get a lot of purpose from it, but I do think if we and I'll just say one sentence on this because it could go down a whole rabbit hole, but I do think if we do universal basic income or something, I think that is a good thought.
>> I I think it's a good answer to this >> and I do think that's a good thing if people have, you know, the very controversial very controversial >> but if they have their fundamental needs met, they can do whatever they want.
garden art, play video games, whatever the [ __ ] they want.
>> Where's the money coming from?
>> The money is coming from Open AI, Anthropic, all these AI labs that are just soaking up all the value.
>> Do you think that those CEOs though want to just give away that much money?
>> Proposed Sam Alman, the CEO of Open AI has proposed this. And I think a lot of people that are heavy into AI see the huge problem and they are for it because they understand how big of a deal it is >> because it's it's really difficult to avoid cuz how can why hire someone less efficient?
>> Sure.
>> Exactly.
>> Exactly.
>> He doesn't show up on time sometimes. He fakes he's sick. Yeah.
>> Yeah. I got the flu today.
>> Get sick, man. They do not, you know, >> and he does a worse job at his job.
>> Yeah. Yeah. You know, >> well, uh, this is something. Does it really just take one Sam Alman company to be able to pay 350 million Americans?
>> I mean, your question is, how do we finance this? Right. And that's always the big question with this. And I think that's that's that is the question.
>> Even if him at a press conference is like, "Oh, yeah, I'm willing to do it."
Doesn't mean he can feasibly do it or is actually willing to do it.
>> I I don't know. That's above my pay grade, but >> but these are the great questions to ask. Yeah.
>> Yeah.
>> Let's shift gears here, guys, for the final segment of the first episode of Nocturnal.
>> I wanted to deep dive in as you're looking at a logo up here. You see all the stars back? There might be a UFO or something back there. Sula, you have one of the greatest minds in my short time of knowing you in my opinion of the sci-fi exploratory anything is possible type of mind, which is my type of mind.
>> Yeah, man.
>> And you had some you had some of the most thoughtprovoking thoughts on, you know, just like humankind. What if we're first? What if we're last?
>> Right. So just having an open mind, you go down all these different rabbit holes. You know, you wonder, you know, what else is out there? Like mathematically, it makes sense for us not to be the only intelligence out there.
>> Agreed. Yeah.
>> But it is also possible that we're early.
>> Sure, >> it is very possible there's, you know, that that we're early and then we go off to populate the rest of the universe.
>> And early means we're the first intelligent species.
>> Yeah. to evolve and survive to this point.
>> And it would make sense because if there were any other, you know, intelligent species out there that were even, you know, just slightly older than us, you know what I mean? They we would have something there that would signal that they're out there. They would have found us already.
>> Let me, you know, say the following things.
>> Yeah, I have thought a lot about this.
>> A lot about this. And I have a few things to say. Number one, >> the amount of science that would go into an intelligent species not only finding us traveling here and being willing to come down to our planet. Here are the things amongst a thousand others.
>> The gravity of their planet. If the gravity is even a little bit off, not only do they need a special space shoot, a space suit >> to be able to walk on our planet, they need to like reinvent how they are going to they can't just land on the White House lawn. If the atmosphere is a little bit off, if the radiation is a little a microscopic bit off, >> our our biosphere here that we all evolved in, our insects, our plants, everything, algae, mushrooms, >> that is all potentially poison to them and will kill them immediately. So then another reason they can't just land here and walk out on the White House lawn so to speak.
>> If on top of all this, if the rate of speed of our planet rotating is not a speed that their species evolved with, then they're going to be ripped apart atom by atom if they even land here.
Mhm.
>> If the you can go down the line here, there are 10,000 physics things. The planet has to be so perfect, so godamn perfect for them to even be willing to consider landing here, exiting the spaceship, and walking on our planet, even if they have high tech space suits. So my >> it's it's it's it's one in a billion that that stuff lines up, >> you know. So well my thoughts is just >> does not mean they don't exist.
>> The radio satellites and all the things that we have out there has detected pretty much nothing burger >> SETI.
>> So go ahead.
>> It's called SETI, right? SCTI. Hey, don't get me started on the search for extraterrestrial intelligence SEI because according to even NASA themselves, >> SETI has only >> scanned one grain of sand in the entire ocean and they think it was a failed program from the start from their own voices.
>> There's lots of debate around that, too.
People think it's a horrible idea to broadcast signals. No, no. The horrible idea was thinking that we're going to use radio technology and that any intelligent species automatically also has archaic radio waves.
>> Are you guys still transmitting radio like 10 billion years from now? Right?
Come on. They're not still transmitting radio. We're not even transmitting radio anymore. Like, come on. Give me a break.
>> And it's also assuming that they invented the same technologies that we have.
>> Exactly. It could be a different entire tree of technology.
>> It's, you know, a unfathomable thought process. You know what I'm saying?
Something, >> you know, that we can't even fully comprehend. You know, their concept of language, concept of everything could be completely different if they exist. But what makes me really lean towards the first theory is, you know, more of the things that we're seeing now and the capabilities that we're starting to see happen and just no evidence uh so far of there being anything out there. Uh it just makes me lean towards it. I I play around with that idea a lot because and I you know study a lot of other great minds and they all kind of think similar and and and some people think different you know we so just like you said all the other you know >> like other places China these other like if we really start putting all these resources Yeah. into populating or traveling, you know, to other planets and and putting bases there and things of that. What Well, what what could happen, you know, is the other, you know, places >> I like where this is going.
>> Start to compete.
>> Yeah, exactly.
>> And that competition can turn catastrophic.
>> It sure can.
>> You know, and um I mean that's just going down a whole rabbit hole. Well, then you then you got to start wondering, you know, uh Elon Musk, he's going to start pumping out these rockets at these factories a thousand per year is is what they're aiming for. And we're going to be populating Mars before we all know it for real within the next couple years. And so then you have to understand, we all know this. This is this is not up for this is science that you you have enough people living in an environment for a long enough time, they're going to start genetically adapting traits and they're going to have their own >> Exactly.
>> Right. So then they're going to be the Mars people. Maybe maybe their pigment or something adapts to the radiation and stuff from the sun out there at Mars.
Maybe maybe their lungs or something with the dust in there.
>> I'm saying all that to say like modernday Star Wars.
>> I'm saying it feels like we're at the beginning of our Earth is like the ancient, you know, pro, you know, planet. You know what I mean? We're we're freaking like the beginning.
>> It feels It feels like this is the most like cornerstone threshold moment of all of human history. It's the AI wave. It's the space wave.
>> And how did we get to be born during this time?
>> As creators, I think, >> hey, this is what Nocturnal is all about is to talk about this [ __ ] at 3:00 a.m.
in the Midwest. AI, aliens. This is it, man. This is the show. This is it.
>> Get her done.
>> Um, have you guys heard of the Fermy paradox?
>> Yeah, of course. This is my This is my worst enemy. So every >> my worst enemy is this.
>> Everything we're talking about is essentially the crutch of the firmy paradox.
>> Dude, don't don't do this.
>> I think it's we'll go down it just a little bit >> because I printed literally a lot of paper. Go ahead. Go ahead.
>> So the the concept of the Firmeny paradox is that as big as the universe is, >> we're talking infinitely big, that there's got to be other life forms out there. There's got to be other planets that are similar to Earth that are as close to uh their own sun and life life form.
>> It's called the Goldilock zone.
>> The Goldilock zone and the universe is two billion years old.
>> We've made our scientific and technological progress in a few hundred years.
>> Mhm.
>> So in a two billion in a two billion uh year um time span, 200 years is a [ __ ] blip.
>> Yeah. So the whole concept of the Firmeny paradox is if the universe is huge and there's an infinite amount of other life forms out there, it's been around for two billion years, where where are they? Where are all these civilizations that should be advanced?
Should should be out there, should be visiting us or sending signals. Where are they?
>> That's the crux of the question. Like where are they? And you know, Sulie, you've mentioned a lot like are we first? Like that's one of the answers to the um Fermy paradox that life takes a long time to form. intelligent life >> and so many things have to go right.
>> So many things have to go right and we just our humanity just happens to be first and at the first and forefront.
>> Great question.
>> You know there's other there's another um uh idea called the great filter >> and the idea is that at some point a technology advances so much and maybe they all get nuclear warheads and they blow everything up and like you just can't get through this filter where you you get to this technology that can >> professor the infinite loop.
>> So the whole point is it's a paradox.
It's like the the universe is big. The universe has been around for two billion years. There should be all these advanced societies. Where the hell are they? So, where the [ __ ] are they?
>> Well, let me put it to you like this, >> and this is something I've been very passionate about for a long time. And I started to get in there. I didn't expect I was going to have to get in the weeds here, but let me give it to you like this. When you ask the question, where are they? What you're really asking is, do they exist? And do do we have do they want us to know that they exist? I think that's really what it comes down to.
Because when we go to the moon or Mars, even though we know they're barren planets, what are we doing? We're sending nonhuman robots and orbiters.
We're basically collecting science. And the number one rule of NASA is noncontamination.
All of those labs, they're wearing the full mask. the full hazmat stuff because they don't want to contaminate other planets. All of our satellites are of this extreme safety protocol to not contaminate.
>> Why is that? I've never heard that.
>> This is a real thing. Because we don't because they had the foresight to think, yeah, if there is life, even if it's microbial, we don't want to contaminate interfere, right? Because the whole thing is it's a biosphere thing.
Everything on Earth evolved to work together. Even our flu, our viruses, our bacteria.
>> You have one microbe from another planet. You can eradicate an entire planet overnight because there's no solution in that DNA to solve that thing.
>> Interesting. M so everyone even even the Saturn U mission stuff that's way out there beyond Pluto that stuff even followed the same protocols but you have to think about all the factors does that life have the same atmosphere is it carbon is it you know what's it based on is it what are the gravities what what are the microbes the biosphere the radiation the magnetic field you got to think about sensory overload >> do they even have the same sensory input But is our light radiation just like disease to them? Is our pollution?
>> Would that just like rip theirselves apart? We don't know.
>> There are so many factors. It's not just like, hey, they're just like humans.
They're just going to come down here and be like, "Oh, don't you know?"
>> But as you and as you see but as you see our creations, it really it really just starts to feel like we're, you know, cuz cuz you want to be a humble species. So So initially you want to think like mathematically there has to be other species out there.
Yes. But as you see how things are progressing and how many things have to go right, it seems like we're during our lifetime, we could start to see the Mars people as you're saying.
>> Yeah.
>> And it it could be your cousin.
>> Sure.
>> You know, one of your cousins could be a Martian and you know, you know, and you may see it like you may be 75 and >> he's got red skin because he's been on Mars or something.
>> Is a native Martian. I think that'd be cool.
>> It was born on Mars.
>> I think it's I think it's right around the corner >> and then you and then you find and you know it go it goes back to time >> which is relative >> you. Yeah. So so then you know you can start getting real crazy with it.
>> That's real, >> you know, and wonder. Okay.
>> Well, >> okay. Go ahead.
>> Okay. Hold on. Hold on. Hold hold hold the [ __ ] on. Here's my point with that.
Time is so relative. If your planet if you evolve on a planet and they and and they evolve and the planet rotates at such a speed >> with and and the mass of that planet is is so high.
>> We've seen interstellar.
>> No, but what I'm saying is they may visit us and we're either moving in slow motion >> or we're moving so quick it just doesn't line up.
But the other side of this is of course now we have these independents, the SpaceX and stuff, Blue Origin, >> Blue Origin and so forth. And now it's like NASA is being crushed on >> by all just obliterated absolutely [ __ ] obliterated.
>> It's a good thing for the industry.
>> No, it's fantastic. Yeah, >> I think it's the best thing possible because >> it's exciting >> because now now you have companies that are, you know, building this technology where they can they can travel >> and it's not coming from us. It's not government, you know what I'm saying?
>> Yeah. It's entrepreneurs like yourself.
>> You have, you know what I'm saying? Go ahead. I think I think one of the the byproducts of this that's going to be really cool, this will be commercialized, all this private space stuff going on, >> but it kind of it's starting to >> it's commercialized with Starlink right now. um kind of satellite technologies and stuff. But the the what I'm getting at here, the cool part is they will start doing commercial space flights in our lifetime certainly.
>> And it'll be a tourist thing. It'll be something interesting to do.
>> They're kind of doing it, but only for billionaires.
>> They're doing it for Exactly. But you know how that technology goes >> billionaires and then it just goes down.
So I don't know how many years, but like we will be able to be alive >> $1,000 of today's money or $2,000 to go take a trip in orbit or something. Well, I'd prefer I would prefer going to the Mars Resort.
>> There goes your ROI right there.
>> Now it makes sense.
>> Well, can Okay, can we all just agree at Nocturnal, we're all going to take a group trip to the resort on Mars? Yes, dude. Yeah. Hell yeah. Yes. 1 million.
>> We might be We might be old men at that point, though.
>> Yeah. Well, >> maybe >> the Mars the Mars one. We might be old man.
>> We might be able to pick an avatar that looks like our age right now and transfer our consciousness. Hey Now you're talking. You're just the singularity is near.
>> Yeah.
>> We are in a society where >> there's a lot of closed minds.
>> It's so polarized. It's so black and white. Everyone is convinced their team's right, their sides right.
>> The world would be so much better if the minds were wide open if people lived in that gray area because that's where the truth is.
>> The truth is in the gray area.
>> It's in there, man.
>> You got some truth to what you're saying over here. You got some truth to what you're saying over here. That does not mean you are correct.
>> You know what I'm saying?
>> Yes, exactly.
>> It's been a great show. My name is the double Ollie Olsen. On my left is the one and only Sulie 4Q.
>> Mhm.
>> There he is. And of course, Justin right there. Thank you so much for watching the first episode of Nocturnal. Don't go anywhere. Get the [ __ ] out of here. Good night. Bye. Time.
Videos Relacionados
OpenHuman VS Hermes AI: Who Wins?
JulianGoldieSEO
285 views•2026-05-29
BREAKING: Microsoft’s New Image Generating Model Beat Out GPT 1.5 and Nano Banana 2
aimmediahouse
122 views•2026-06-03
Long-Running Agents — Build an Agent That Never Forgets with Google ADK
suryakunju
142 views•2026-05-30
This computer is made from real human brain cells. And you can buy it.
Talktmsmedia
3K views•2026-05-28
I Made the Same Anime Fight Scene in Every AI Video Generator
NobleGooseAnime
295 views•2026-05-30
Nvidia Bets Big On AI PCs | New Chip To Power Windows Laptops | Technology | AI Updates | N18S
cnnnews18
3K views•2026-06-01
I Tested NEW Opus 4.8 on Four Projects (Updated LLM Leaderboard)
AICodingDaily
298 views•2026-05-29
3D Platformer Update - NO CAPES
SolarLune
294 views•2026-05-30











