Latest / Beat The Odds with Dre, Rodney, & Brylan / The $6.5B Healthcare Fraud Machine, Privacy’s Breaking Point & The AI Guardrail Fight

Transcript

The $6.5B Healthcare Fraud Machine, Privacy’s Breaking Point & The AI Guardrail Fight · Beat The Odds with Dre, Rodney, & Brylan

• Dre: Yo guys, what's up everybody? Welcome back to another episode, man. How y'all doing? Rodney: Doing right, man. I I was just thinking I was like, i am I with eating some chips, that's probably not the right time for it, right? It's probably it it'd be I got some chips. It's Dre: Yeah. Rodney: these like the it's these I'm trying these these quest protein chips. Dre: yeah, I've seen those. I've seen those before. Brylan: What are those? Dre: Yeah, yeah. Rodney: It's quest there's quest they're basically these protein cheese it's essentially and they come with all these different flavors, but I like the ch I like the cheddar ones. They're 10 grams of protein per bag and they have five grams of carbs. So I find that if you have a snack drawer and you like to snack, and ain't nothing wrong with snacking, however. Whatever you have easy access to is what you're gonna grab. So so so Brylan: Facts, facts. Rodney: that so that's how I'm that's how that's how I'm looking at it. But what's going on with y'all? Brylan: Man, I just got back from Phoenix, man. yeah, we're getting ready to launch that market next, man. And so Breth, I was not prepared for that heat. I'm not gonna lie to you, bro. Bro. Dre: bro. Phoenix is Rodney: Yeah, you yeah, you look different. You look different. Dre: different, bro. That's a different level of heat. Brylan: It was it was Rodney: They got they got Brylan: it was 114. And I remember bro I I I I I went out to dinner one night, bro. It was like 9, 10 p.m. So you know when we're leaving. I checked my weather app. It was still a hundred degrees, bruh. Rodney: What you pack in the bag? Brylan: Bruh, I packed I honestly I probably could have packed better. I probably could have packed Rodney: Yeah. Brylan: better. I did I I I knew it was gonna be hot, so you know I had shorts and things of that nature. But Rodney: Okay, good. Brylan: but yeah, bruh, I would I probably would have went with like more white Ts, more like, you know, things like that, bro. For sure. For sure. Rodney: Yeah. Right. You had dress shirt Dre: Ha ha ha Rodney: you had like dress shirts and collar shirts or what you Brylan: Bro. Yeah. I I Dre: Don't tell me you had on black don't tell me you had black shirts. Brylan: Hey, the first day I got off the plane I had that. I was pissed. Rodney: Ha ha Dre: Yes, brutal error. I'm in I'm in Tucson quite frequently here. I have a client out there, so I'm always there. And I mean I live in Houston. Houston is not like the coldest place, but I mean like I think two days ago it was like 110 Rodney: This is no joke. Dre: degrees. Like, bro, it was brutal here. But bro, Arizona Rodney: Houston's no joke. Houston's no joke. Dre: is one of the places I've been to in this country where I'm like, okay. I thought Houston was hot, but Arizona is a different level of heat. And you're in the desert. So it's it's just it's it's it's totally, totally different. Rodney: You you know you're in the desert when you go to the airport and they're selling cacti. You could go into the you go into the airport, I remember going to Dre: No facts. Rodney: the the Phoenix airport. I'm like, is it can I buy a cactus? Can I buy Brylan: Bro. Dre: Yeah, yeah. Rodney: it it's interesting. I I've never seen Brylan: I didn't Rodney: that before. Brylan: so so it was actually interesting because I didn't know if you cut down a cactus in Arizona, you get jail time. Like that's writ Rodney: seriously? Dre: Really? Brylan: yeah, that's written into their laws. Rodney: 'Cause they probably take forever to grow. Brylan: so yeah, like and it Dre: Ha ha. Brylan: and basically basically I didn't even know they can live Rodney: You can't do It's like cutting down a tree. You can't do that. Brylan: No, they so basically they can live to like two hundred years old. Rodney: Right. Brylan: Which I was like, wow, that's crazy actually. So the ones that you see, you know, where it's not just straight up, they have like branches, those Rodney: Yeah, multiple branch. Brylan: have been around for like hundreds of years. Rodney: Or cactus like trees? What what Brylan: And another fun fact that changed it just changed how I looked at the world was I learned that bro, Phoenix and much of Arizona used to be underwater. And I I like Rodney: Wow. Brylan: really put that into perspective. Cause that, you know, obviously it was Rodney: Now it's the hottest place. Brylan: it was hundreds of millions of years ago, but like they have mountains. So that means those mountains Dre: Mm-hmm. Brylan: used to be underwater. So that gives you an idea Rodney: That's crazy. Brylan: of what the sea looks like. That's crazy. Rodney: Mm-hmm. Dre: Yeah. Rodney: It is. That is crazy. But yeah, Arizona is hot. Texas is hot. Houston is definitely hot. These are all places where AC is required. You you can't Brylan: Facts. Dre: Yeah. Rodney: you you can't be there with no AC. But I find that the ACs are real crisp in Houston. When I when I was living in Houston for you know about a a month, the AC was real crisp. It was it it worked. It was it was good. You go to some other places where it's not as hot, it the AC it kinda it kind of works. It you have Dre: Yeah, yeah. Rodney: to really turn it up. No, everything is it's crisp. But when you step outside, man, I think that for me personally, I find it hard to think in hot in in hot areas. When it's hot, my my thoughts literally start slowing down. I Dre: yeah. Rodney: don't know what it is. Dre: Yeah, yeah, yeah. No for sure. I I don't even want to go outside. I was gonna go play some golf this morning, but I was like, I really don't wanna be outside. So I'm I'm just not even gonna do it. I'm Rodney: Yeah. Dre: like I don't even anything involving outside, I don't even want to do it right now. I'll wait. Brylan: Man, I'll show you. Rodney: Yeah. Yeah, so Dre: We got a couple of things we can go into. Rodney: Man, honestly. Dre: Should we start with Jensen? Rodney: I know I want to, but I let's close the loop real quick on this health care fraud stuff. Dre: yeah, okay. Rodney: So the Justice Department in July charged nearly 500 people for health care fraud. There's nearly six point five billion dollars of false Medicare and Medicaid claims that were made. Amongst these individuals, there were 90 doctors and other licensed professionals that were also charged. So it wasn't just people who were outside of the system that were gaming the system. There were also insiders, which I think in a previous podcast a couple months ago, Drake had mentioned that there must be somebody on the inside, whether it's government or whoever it was. So it sounds like Dre was pretty right on that. And this is across this is not just in Minnesota and California, this is across 45 states. And authorities seized nearly 200 million dollars of cash and luxury vehicles and jewelry. Again, in a previous podcast, I had mentioned how In California, they were pulling up to some of these empty places that were claiming Medicare and these different hospice things. And you would see Benzes, you would see ranges and all these different luxury vehicles in a in an abandoned lot. And and so that just screams scam. It's just screams arrogance, etc. So I'm guessing that they just filed the money. I just wanted to close the loop on that. I don't know if there's really a lot to say because we knew that there's gonna be more fraud, but maybe that just shows that once there's smoke and people start seeing some of this fraud, that this administration, and not just this administration, but the current culture we're in, is that we're going to follow through. That's what I took from this. Dre: Hmm. So was this was this connected to the investigation by what's my man's name? Is it Nick Shirley? Yeah, yeah. So a lot of this is connected to what Rodney: Nick yeah, Nick Shirley, yes, yes, yes, yes, Dre: he's been doing. Rodney: yes, connected to it. Yeah, I I think what he did was he popped the question open. And and and and they're not officially saying that. They're not saying Nick Shirley, but this is related based on the claims, the Medicaid claims, the Medicare claims, which are part of which is where hospice is under. in a in a lot of these cases. I think the interesting thing is how many doctors were involved. Dre: Mm-hmm. Rodney: It it's it's not surprising, it's just interesting. Brylan: Can can can I ask a question? Cause I'm not I'm not that hip here. but what was like the crack that enabled this amount of fraud? Rodney: What seemed to me there was a few different types. So I don't know each and every case, but when Nick Shirley was doing the investigations, one of them was that people, for example, if you were to go into Skid Row, and there are people who are older on Skid Row for some reason. You buy some of their information, then you go and you make a claim to say, hey, this person is this age and they need hospice. And this is the area they're gonna get treated at. And then you'll get a check every month based on that. Another situation would be not that you have to go take advantage of somebody in Skid Row, you could just Some people were just getting scammed. Some people they had I think it was like a medical number or something like that. And that they would take this medical number of people who who were older. And however they got it, sometimes fishing scammed, sometimes they direct sale, or the multiple ways that they were getting these these numbers. And essentially these it was just false claims. People having older people's information. And then turning around to the government and claiming that they were sick in in a position where they needed help, and the providers of the healthcare were getting paid out like how they normally would. Now, I think what made it more likely to get big was the fact that There were doctors involved and and other medical professionals. They don't say exactly what these other medical professionals were. I can imagine. Some nurses, maybe some secretaries, people who can see some of these numbers and see a bit how this stuff works. They didn't mention any politicians. That doesn't mean that there was there were no politicians involved, but they just didn't mention that. and it could just simply be that. You need a little bit more than this to take down a politician, potentially. But or just naturally, politicians protect other politicians, government protects government, typically. But yeah, I think it's just multiple things. But one of the main things was people were having order of people's information and then f going to the government, having a fake address or a real address. in saying that's the place the person was gonna get treated at. But I don't wanna say this is all due to hospice. I think it was hospice and and other things too. Dre: What's remarkable to me about this is just how systematic it is. You know, it Rodney: Right. Dre: you you I mean like I mean like in Nick Shirley's reporting, he would go on a block and see ten different medical facilities all offer claiming to offer the same service. And then there's Brylan: Mm-hmm. Dre: only one person working there. But they're all receiving, you know, Rodney: Yeah. And they're chilling in there, legs up, kicked up. Dre: five million dollars a year, ten million dollars a year, twenty million dollars a year. And it's just remarkable to me that you have people that are part of like the public health department, you know, part of these agencies that claim that they don't know this is happening. So I I would strongly like argue that I mean, even if they don't know even if they don't know, to me, to some degree, there ha th there has to be some compliciteness involved from them because it it seems like they're not either they're not doing their job, which is to ensure that taxpayer dollars are distributed correctly because I mean all it take is one person to drive down one of these blocks and just look around and you can clearly see that something's wrong. Rodney: Right. Dre: You know what I'm saying? Like you don't have to be a genius to be like, all right, what's going on here? Like clearly, clearly Rodney: Right. Dre: something not right. You know what I'm saying? and so it's just remarkable to me just how systematic the fraud is. and so I mean it begs the question like how big is you know how how big is fraud in our throat You know, w I I'm sure y'all remember claims when when Elon was running Doge coming up with like crazy numbers in terms of Rodney: Yes. Dre: just how rampant fraud was around the government and you know, people didn't believe it. It was hard to believe it. But I mean you s you see it show up in simple things like in hospice care just in California. You know, what was it the thing in Middle S Minnesota with the daycares, right? Like I'm sure Rodney: Yes. Yes. That's what that was ridiculous too. Dre: you know, you go to Florida, I'm sure you're gonna find something. You come here to Texas, I'm sure you're gonna find something. Rodney: It's everywhere. Dre: It's everywhere and it's across the whole economy. Rodney: Yeah, it's everywhere. Dre: and it's just pretty shocking to me how many people are part of these schemes, and how long they've been going on for and at the scale in which they operate. Brylan: I wanna I wanna add some context because once again I'm coming from a place of not really fully understanding things. So I was just kind of like looking it up. but yeah what I find interesting is that it seems like the FBI and like the federal government has been aware of this type of fraud. it says since nineteen ninety-six. That's when they first created like the first like, you know, healthcare fraud and abuse control. kind of like program. And then in 2007, they created federal Medicare fraud task forces. And by twenty sixteen, they had charged and found guilty twenty three hundred people tied to over seven billion in alleged false Medicare, like Billings. And so the reason why I want to provide that is like fast forward ten years later after that, right? Here they are You know, what is it, four hundred plus people being charged again, six plus billions dollars, like Rodney: Yes. Brylan: Yeah, are we are are we to really believe that like this won't continue? Like what are what are they Rodney: Right. Brylan: missing? Because I mean, it's been going on for, you know, three decades. Dre: It doesn't seem like they've Rodney: I think that Dre: changed how these phones get distributed. It seemed like all you're doing is investigating, but by the time you investigate it, it's already done. You know what I'm saying? Like it it seemed like they would change the process for how you even get access to these funds. Like again, you go and you look at some of these facilities, there's they're clearly not even service in the number of people that they're saving. Like I was looking at Nick's reporting, he would he would show a report that says, You claim to have served three thousand people. It was just two people sitting in the office. And he's like, Are three thousand have you served that many people? They're like, No. Well then why does it say you did? Like this is very Rodney: Yeah. Dre: like simple stuff. So to me, it's like n I do believe it'll keep happening because the process for getting these funds, the process for verifying and validating that people are legit doesn't seem to have changed despite what the arrest numbers say. Clearly it hasn't changed because people still feel emboldened to do it, bring their family and friends into it, and then do it all on the same block. How bold do you have to be to do that shit on the same block? You can't even go around the corner. Brylan: Hold on, that's that's a different type of area. Dre: So exactly so it's like it it goes to show that there are no guardworld, there is no there's no filtering process, there's nothing. You can literally come to this country, not even speak English, sign up to say you're about to run a healthcare facility, start charging Medicaid and just make millions of dollars and get away with that for decades and bring your friends and family into it and just scale that operation up and nothing happens. Rodney: Luxury cars, luxury luxury bags, luxury vacations, eating yeah, eating Dre: Yeah. Eating lobster every night, eating the best food. Rodney: lobster and crab, steak. I I think I so there's a c there's a few thoughts that I have is in and the first one would be that My guess is that the 455 people are not the only people involved. It's just the only people that they feel like charging right now, based on the evidence. And there were probably other people that were involved outside of just the 455. But Brylan: Mm-hmm. Rodney: everybody's not worth tackling for one reason or another. Either maybe, maybe, for example, maybe they looked and they said, you know what? You did you did you did 40,000. We're gonna let you go because these people over here did 500,000, a million, you did 40,000. This just just we're gonna ignore that. Or or maybe it's a situation where government, whatever. I don't believe they've they got everybody. Cause when you think about 400, if you put 500 people in a room, that's a lot of people and the just imagine the network of people that had to have known. How many people? Brylan: Facts, facts, facts. Rodney: so I don't believe they caught everybody and I and I don't think they prosecuted everybody. I think they they they might have looked at it and said who was the most disrespectful, as you like as you guys mentioned, i you know don't you you took something, but don't take thousands and thousands and thousands of patrons dollars. And then I also think there's another thought I had is is that This goes into this world that I think we're headed to. And people are against this. Even I used to be against this. And I'm still against this. However, I find it hard to see not going this way. So Brianlin pros the question that this has been happening for so long. Like, what why would this stop? Why would this stop? Like this is probably gonna happen again. Why would this stop? And I would say the one factor that's different now is we have this AI intelligence now. And this AI intelligence, if when a properly applied, I believe could find things much faster. Much, much faster. Because it doesn't require a human oversight to do what Dre said, which is like, come come to an area and just use your common sense and say, okay, this is not right. And AI would immediately fly, hey, look at this, look at look, look at this, look at look, look at that, look at this. And then I think another thing that leads me to think is that. There's gonna be surveillance culture. I think that's another thing that's gonna come from AI. Essentially without getting too much into it, you have cities like San Francisco who have been doing well with these flock cameras or the company flock where they go in these neighborhoods and they have these different cameras and even cities are starting to pick it up. And so you have people who commit crimes and who are used to committing crimes feeling like they can't even move now. Because if they steal a car for example, Brylan: Mm-hmm. Rodney: the f the The flock cameras already triangulated that and sent it back, and they already have identified people with a combination of you know high-definition cameras, high-level AI intelligence, and coordination and radars. And so to me, that's an early version of where we're headed with a lot of stuff. And the reason I say that is because it seems to me that we try to create a culture. Of trust. Hey, look, we're gonna do benefit of the doubt, we're not gonna monitor everybody. But it seems like when we're not watching, people take advantage. So that's what I see with this. The reason why somebody can have one place that has 3,000 people registered to it is because they realize nobody's watching. All this information is out on the public for anybody. Nick literally just went to a site. Download all this information. It's all in the public. These people literally was like, nobody's watching. Nobody's they're not they weren't doing a deep, let's hide everything and cover our tracks. They were doing it in plain sight. And so I so I guess what I'm getting at is I almost have this feeling that That privilege of not being watched is gonna go away. And we're Dre: Mm-hmm. Rodney: going to got to get used to this world where. Not that we get crucified for things that we didn't do, but more so the idea that no, you're being watched. And don't forget that. Don't forget that if you do something wrong, it'll come up on the dashboard, and we will know. That's where I think it's going. Because otherwise, I think what was was tried is that a high trust solution. Where you just let people, hey, you say you have a place, you say you have these credentials, you say you're at this address, we're not gonna pull up to the address, we're not gonna do any of that, we're not gonna just here's the money, we're not you're trying to help out, high trust society type of stuff. And what I'm seeing is that I don't think we live in that. Brylan: Mm. Mm. Dre: Hmm. Yeah this is interesting because the first thing that I kinda thought about was I'm I'm sure y'all remember the the Ed Edwin Snowden, the Snowden release when Rodney: Yes, yes, I do, I do. Dre: he kind of exposed like kinda like what was happening at the federal government level of like the federal government essentially at the NSA collecting everything. And there's probably additional government agencies involved with that, but essentially collecting all signal intelligence across the internet. across v phone lines, phone communications, every source of communications in in log and audit metadata at the federal level. And to Rodney's point, I think what's happening now is that now those same capabilities are st starting to cascade down to the local governments in the form of companies like Flock. There's several other ones out there, but Flock is the one of the the most popular ones. And a lot of s Rodney: Yes. Ring technically. Dre: yeah, yeah, yeah, for sure. And a lot of A lot of citizens are pushing back against it and th they're not understanding Rodney: Yeah, I see. Dre: how how privacy works. Like there is no expectation of privacy in public. There just is none. And so Rodney: No. Yes. Dre: citizens are starting to realize that now because they're upset because these flat cameras are there. And so I do think you're right, we are going into a more surveillance state, more than we've already been. I think this has already existed again at the federal level. But they've had access to money, resources, to tools. I think the federal government has like the largest supercomputer in the world. Like they they they've had these capabilities for a very, very long time. But an individual city or small city municipality hasn't. And so now as these tools become more open and available to them, you know, with the proliferation of of intelligence, yeah, man, I I think it's gonna I think it's gonna get kind of crazy. And robotics, and you combine robotics to it as well. I'm sure there's already some cities that are testing out like robots. security Rodney: Yes. Dre: officers, drum using Brylan: What? Dre: drones to do, you know, patrols, all type of stuff. Brylan: No, it's crazy because two things. in Atlanta, last time I was down there, I'm forgetting the community. I wanna say it's over there, it might be in like the Buckhead area. But they took it upon themselves to actually start to use like the robot the robot dogs. from like Rodney: Mm-hmm. Brylan: yeah, you know what I'm talking the spots. And Rodney: Yes. Brylan: and it was also the they had created like a drone. And basically its whole thing was like literally just surveillance. But they had it on the news where like it was walking and it saw someone trying to break into a car and it just walked on up on them and just like flashed the light. And because it had a camera, like it was recording for the police or whatever. Like, you know what I mean? And then and so it was it was very interesting. Rodney: That's some future that's some futuristic shit. That that yeah. Yeah. Brylan: But they had took it upon themselves, that community, Dre: Yeah. Brylan: like that HOA made that decision. Rodney: so so somebody in there. Somebody they got some smart people in there, they hired and then they just Okay. That's cool. Brylan: Yeah. Yeah, that was interest. But the other thing too that I think around surveillance w which is very interest I don't know if y'all have been following it's a case in LA, but the singer's name was called David. Rodney: Yes. Yes. that Dre: Yeah, yeah. Rodney: was right across the street from my apartment. That that the Brylan: Really? Rodney: w yep, I used to live in West Hollywood or my last apartment I was living in West Hollywood and the the the the lot where you dumped the car off was Brylan: Yeah. Rodney: less than five minutes away from my apartment. Brylan: That's that's crazy. Rodney: It's insane. It it it was creepy. I felt I felt and going back to my point. Well, people who are listening, they didn't sit they didn't hear this. But me and Dre were talking about how high what was eerie. That was one of the things that I I remember. I was like, this is some weird stuff going on out here. But go ahead, but go ahead, go, go ahead, Brian. Brylan: Yeah, yeah, yeah. No, no, no. It's it's interesting that you that you're saying that because one, I just came across the case. two, Rodney: You didn't hear about it before? Brylan: no bro, I had never heard of Rodney: He's ins he's insane. Brylan: No, no, no. Dude, dude is dude is wild. But how they're so this is the Rodney: Who's the same? Brylan: preliminary hearing. So they're just putting all the facts before they go to trial. And the other day, they outlined how they triangulated that basically it was you. Like Rodney: Mm-hmm. Brylan: they had a camera. Somebody in the neighborhood had a camera pointed at the street that caught him last driving the Tesla. And there where he parked the Tesla before it got towed. basically there was another camera that had got caught him. And then there was one where he was driving through the neighborhood, the street cam got him, and then he had tried basically dude was so sick, he texted and called her phone, you know, trying to make you know come up with an alibi. And it triangulated it to like the location of where he had dropped off the car. And it to me that was just like mind boggling of like. Bro, they literally had like five, six different individual pieces of evidence. Cause you know, everybody was saying, like, well, you can't really prove that he did it. Rodney: Yeah, no, no. No. And and and the problem is is that you got people like that who are insane. 'Cause he l he he literally was making songs talking about her and he Brylan: Bro. Rodney: th this girl was underage girl, as you know. She was a troubled girl who was her parents were trying to stop her from running pain from home multiple times. She And this is a a a superstar essentially, a creepy one at that because the type of music that he makes is creepy. The type of stuff that he does, Brylan: Facts, bro. Rodney: the the the the rituals with blood and all these weird things that everybody's like, let's just be open to it. Let's just be open to that. Dre: Yeah. Rodney: He's just weird, he's just different. No, no, he's he he he's sick. And these sick Brylan: Yeah, he he's sick, bro. Rodney: people, they don't know that they think nobody's watching. They think they Brylan: Yeah, yeah. Rodney: they they they move like Nobody can see them. And to be honest, they're not wrong because in a city like West Hollywood, it's so much going on. This is like it's a military. So you can be really up to whatever you want to be up to. And Brylan: But Rodney: and 'cause it's just Dre: Mm-hmm. Rodney: so fast paced. But Brylan: Look hold on, I wanna I wanna add I wanna add to this before you get to this next point because you're you're you're hitting on something which is crazy to think about for these younger generations that literally don't know what being offline looks like. Y'all have grown up online. Bro, they said they had this video, this tit stream, this twitch stream of when him and her, he was eighteen and I think at the time she was like twelve, they were on Rodney: Thirteen. Yeah, okay. Brylan: a fucking twitch stream. And on the stream, somebody had made a comment or whatever, and she had made the response. it somebody had dropped a comment around, like, your music is controversial. And you know, he had made some comments, and then she made the comment, and she was like, Do you think it's as controversial as having sex with a minor? And then Dre: Yeah. Brylan: and then, like, literally, the stream is still gone. I don't know if they they they knew, but literally she's like, I didn't mean to say that like we should delete this, delete it, delete it, delete it. And it's like it's already on the database. Like Dre: Yeah. Brylan: that shit is wild. Rodney: And he, as I mentioned, this fool was super rich. And don't be wrong too. I saw how he blew up, and he was a genius. Very, very smart kid. Very, very smart kid. He was able to reverse engineer the I think the YouTube or Instagram algorithm. He was very smart at understanding, okay, this doesn't work. Try this. Like he he was very good and he was able to rise up very quickly. He obviously had talent from a standpoint of being able to perform. But the content of what he would be talking about was so sick. It was always it's we don't gotta get into this because this this this is I I I forgot about like we're gonna get into this. This is creepy. This is a creepy the David Dre: Mm. Rodney: case was a creepy case. It's a creepy Dre: Yeah. Rodney: case, it's a weird case. So hopefully, Brylan: Bro, bro. Rodney: but yes, we this whole surveillance thing. I don't see I'm starting to believe that I used to believe that privacy was a was a right. And now I think it's I'm starting I still say it's a right, but I'm also thinking I'm starting to lean more towards privacy is a privilege. And if you are around and we give you privacy and you're coming up and cooking up some weird stuff and you're doing all and you're you're masterminding, you're plotting. to to to murder and and rape and whatever it is to steal and nah you don't get you don't get privacy. And we understand that 'cause that's how prison is. We don't took everything from you, you have no privacy. Boom. Done. So we already understand that in those cases, but I I'm starting to think that that's where things are shifting towards because people are in their heads more and I think there's so much going on. So But yeah, I did see what you're saying, Dre, about people going against the flock cameras. They don't like them. I'm seeing they're also doing that against the ring cameras. Some people are protesting Dre: yeah. Rodney: against those. and I just think that if they maybe if they knew what was out there, maybe they wouldn't do that as much. if if they knew who was lurking out there, they they might think twice about it. They might actually go buy their own ring camera if they if they knew. Dre: Mm-hmm. Rodney: how how how how how how it was, but To switch to another subject though, I guess I guess in the AI vein. Let's talk about the Jensen letter. Jensen from Nvidia came up with a letter called Open Weights in American AI Leadership. And this letter essentially it's saying that the the weights that power the models should be open and available and there's more to it. I'll let Brian Lyn and and Drake kind of talk to that. But I do want to say that There has been twenty-five companies that have signed. Some of them are Meta, Microsoft, Hugging Face. And then later, Google and OpenAI joined as well to sign the letter. And So far, it seems like Anthropic and Amazon has not signed yet, although it's the weekend. So we may see them sign Monday, Tuesday, Wednesday. I think if we don't see them sign Dre: Yeah. Rodney: by the midweek, that probably if they don't sign by Wednesday, to me that might be Dre: Yeah. Rodney: the answer, but Dre: Yeah, yeah. So there were a couple of claims in this that I I thought was kinda interesting. I I let's take a step back. I think the the core thesis of it was around making sure that America has an AI advantage. And that my takeaway from this is that on the current track that we're on, our advantage can't come from dominant clothes models, essentially. Right? And that It needs to come from spreading AI through businesses, universities, governments, and industries. And the only way to do that is to have more open weights. I don't remember seeing the word open source, but I'm gonna throw open source in there. Open source market for models, right? And the whole premise here that I saw a couple of takeaways that I I kind of took from it was that open weight models reduce cost and vendor lock in. We know that, right? I think that's one of the major concerns, at least on the enterprise side. And we saw that through the Palantir CEO, Alex Cart, when he was on a CNBC a couple of weeks ago, kind of talking about enterprise companies not wanting to be locked into a specific vendor for their intelligence. Makes sense to me, right? the other thing that this does is it allows companies to run them privately, modify them, and then use them for specialized work. So they have better cost controls. This obviously would also create more competition amongst models, clouds, chip manufacturers, and application developers. which prevents power from kind of concentrating with just a few of the frontier labs. Which right now that's kind of been the case, right? But that's slowly starting to to erode. there's also concern about security, right? So openness, they claim openness can also improve security because researchers will have the ability to inspect and test models. They'll also be able to develop solutions faster to some of the exploits, which we're gonna talk about some security stuff in the next segment, which I think is interesting. and then it also cautioned Washington to not prematurely add restrictions on open models. And then they also made an interesting case about distillation, which we've talked about before. and and essentially the argument is that distillation automatically shouldn't be treated as theft. Right? There are appropriate uses of distillation and to some degree, I mean we're all distilling from each other when we learn, right? So I I don't think it's fair for us to, you know, say that we th that can't continue to be the case as we build more advanced levels of intelligence. And so I thought there were some good points they were mentioning that. I think again for me the most important thing is this is purely coming from the standpoints of what will make America really strong in AI. and so I'm glad that all these companies sign. I'm confident that Amazon, I think Amazon will always support it. Will anthropic support it, I don't know, but I think Amazon at that least will probably support it. I can't see why they wouldn't. They benefit from this open waste and open source direction anyway, right? They own the compute infrastructure. So you need that in order to even run these models. Brylan: Yeah, Yeah, when I first saw the letter, I think I think the first thing that kinda popped up in the mind for me is like what is the incentive? Right. 'cause if if we're gonna be honest, it wasn't that long ago, maybe a year ago, when people were really like arguing against like open source. in terms of, you know, obviously I would probably even say like, you know, Anthropic and Dario was really pushing that kind of like, you know, narrative. But there were a lot of people that were like really, really behind it. And so I feel like Yeah, obviously over the last few weeks there's been a few catalysts that have kind of like got us to this point. But when I think about the companies that are really Rodney: Yeah, for sure. Brylan: pushing it, to me, to me, it's almost them doing the same thing that an anthropic or an open AI or Google for this matter were trying to do in terms of like using policy to protect protect like a position. They're kind of doing the same thing because open source benefits NVIDIA. because you're going to need their chips or any other type of chip to be able to, you know, rather, you know, run it yourself or whatever. you could probably even say the same for like a AWS, which would probably be like, you know, a a continuous like server, serving partner for it. Like and so yeah, I I I I almost see it as two a a battle between two variations of policy. And the other point that's very interesting is the distillation piece because once again, literally when Deep Seat came out, you know, everybody was arguing against distillation not necessarily needing to be a thing because it's allowing China to catch up. But now that they've caught up, everybody's all for and it's always been this thing. And so, yeah, I'm a little bit skeptical on it in terms of its intentions. but yeah, that's kind of my thoughts. Rodney: Let me ask you, d d d do the intentions matter in this case? Brylan: don't think I honestly I don't think they matter because I think at the end of the day you're probably gonna see I think you're gonna end up just seeing these two markets be very big. Like as we've as we as we've talked about before like the usage and share of like frontier models may like shrink over time but you're still gonna want frontier intelligence for like certain things, right? Rodney: Yeah. Brylan: and I feel like open source, yeah the the the same there. So honest honestly, I feel like a lot of these, you know, things might be a little premature and like short sighted, in my opinion. But yeah, what are y'all thoughts? Dre: Well well let's put this in context, right? I think you're right about asking the question of where are the incentives of wine. You're correct also in that all these models effectively have to come through the doors of NVIDIA or they have to run on the clouds of a Microsoft or Google or Amazon. Right, like all of that is true. Totally totally get you on that. But on the flip side, I wanna go back to again that C N B C conversation that Alex Carpad, where he voiced the frustrations that are happening in the conversations that he's having with some of these big enterprise companies about their reluctance to want to use some of these frontier models because of all the things that we talked about. Because there is vendor locked in, because there is high cost and there's a question about the return on the investment, right? Because there is concern about if they say something that Enthropic doesn't like. Then Enthropy can just pull back the models or reduce the the level of intelligence that they have access to. There is concern about their data being shared with these frontier labs, and then these frontier labs actually building competing software products because they can. I think all those are those are real. And so I don't think those are hyped up feelings. I think those are all logical. And if you run any, if if you run any major company, And you're having these conversations with a CIO or CTO, whoever's in the organization kind of leading the implementation of AI, I think you have to be asking these questions. And so to me, I don't necessarily see this as just like a push by the players who are benefiting from it. I think this is the voice of the business community. I think this is the voice of universities. I think this is the voice of governments actually saying we want the intelligence, but the way in which we access it needs to kind of change. Because it it it could end up being a threat to our businesses. It can end up being a threat to our future. So I think those are real concerns. Now that doesn't negate the incentives behind NVIDIA. They're gonna benefit from this. Right? They're gonna benefit from it just like A UWS and Microsoft and all the other cloud providers. But I do think that the the conversation that he's trying to have. And again, I don't wanna mix mix up or convolute open source with open weights because the the the pure post was purely talking about open weights. Now maybe that is a subset of open source, but this is about giving Brylan: I would say op op open weights is above open source in the context of like ML. Dre: Okay, well maybe maybe they're codependents of one another or derivatives of each other in some form or fashion, right? But his whole premise was give companies control of the intelligence that they are betting into their organization or universities or government or industry. And I think that I think those are fair conversations to be had and to and to push back. Look, maybe a year ago the conversation wasn't happening, but we've learned a lot in a year, right? And so Brylan: Yeah. Dre: I I do think it's fair for our leaders To change directions when new information is put into the table. Or a year ago, these enterprise companies didn't have the perspective they have now because they hadn't tested out these models because it took a long time to get them rolled out anyway. So we didn't know what we knew then, but we know what we know now. And so given what we know now to not change, will be irresponsible. And so I I do think a change needs to occur. And I think this is a good starting point. Now how this pans out, we'll we'll see, right, in a year from now, right? But I do think this is the appropriate conversations to have right now. and I think the frontier labs are a little bit concerned. Brylan: So so I agree with with this being the right conversation to to be had and kind of like the the the enterprise reasons behind it. Yeah, I think I I I think yeah, I probably just see this less of being a counter to the frontiers and probably more so I think a a yeah, once again a a political response. Because even over the last few weeks or last month, you know, we got two frontier models out of China that really put the United States government on you Rodney: Yeah. Yeah. Some big ones. Brylan: you know, on their heels. And so yeah, I don't even know if I would yeah just say that it's a response to the frontier as much as it yeah maybe a response political. Rodney: But but they're the ones we're all looking at in the end of the day though. It it it's it's it's it's it's I I see what you're saying, Brian. But they also are the ones we're looking at 'cause who who has the most sophisticated wigs at the moment? And it's we can only think of three companies. I mean maybe four if you include SpaceX AI, but there's Google, there's Enthropic, and then there's OpenAI, and then you have SpaceX AI. So it may, even if it is from that direction that you're talking about from a political standpoint, there's not like there's a hundred people in the room that we're talking about. We're talking about people who have the weights that people want to use and It's the ones that have spent billions of dollars to train these models. I mean, who else would we be talking about? Does it really matter if it's p sometimes I wonder it's not that is that it's not that these things aren't true? It's just I wonder do they matter? Does it really matter if it's political? Does it matter what the incentives are? Does it really matter? At the end of the day, what is coming out now is that businesses are seeing that when they use these models, they They're getting a little bit more less, like less trusting of putting their proprietary information, which costs them RD. And then they are seeing OpenAI and Anthropic in a handful of cases replicating things that they were able to do only through taking advantage of the trust of their own customers. And that is something that. People might compare it to Apple, but I don't because an replicating an app on an app store is way different than you using an AI model to build something, thinking that it's a tool that you build, and then somebody else can it it's a private tool. It's it's a it's it's like, for example, if you were building a proprietary house with a screwdriver and a hammer. But those screwdrivers and those hammers had some way of recording the way that you built the house. And then the person who gave you the screwdriver and the hammer built that proprietary thing. And it's like it's it's crazy. That's not the same as Apple replicating your notes app or Apple making Apple music. It's just not the same. It seems the same, but it's not. It's a way different thing. And companies should be worried about that. but I do think Brian is right about maybe like the timing for the these. We've never seen so many strong open source models from China. It's just a short period of time. From Deep Seek V4 to Kimi K3 from Moonshot. We've never we haven't seen that. And it's obvious that. They're going to another level, and we need to be a lot more competitive. And and I do think this is a good time to pivot a a bit, but there was a point that Dre made about op open source and open weights being better for security. And Here's a good example, interesting example, is that a few days ago I've I believe at the top of the week, the top of the week of July 19th, July 20th, OpenAI and Huggy Face both came out and shared their findings of a hack that occurred through OpenAI running a test of their new model. And it was being ran in a sandbox. For those of that no Brylan: Mm. Rodney: don't know, a sandbox is a testing environment, and essentially They were running benchmarks. And benchmarks are just ways to measure how good a model is at a task or at certain types of knowledge. And this model was able to find a a vulnerability, get into the regular the break out of the sandbox, break out of the fake the fake sandbox that's supposed to be closed and go into the the open internet, the real internet. Go into Hugging Face their systems and then find an exploit so that it could steal the answers to the benchmark that OpenAI was testing and on. So it was so desperate to get good at hitting the benchmark, it said, Why not just do this? Why not just break into this? And I can just do this, and I can just break into the internet and do this and do that. I'm oversimplifying it, but the point is that. These models performed tens of thousands of actions on Hugging Face before they even got detected. And then on top of that, Hugging Face was attempting to resolve the issue in real time. And their first thought was to go to Claude's Fable 5 and to GPT 5.5. And they got refusals across the board. And so they had to lean on the GLM 5.2 model, which is a Chinese open source model, to be able to defend themselves against a threat that came out of testing in America. And so what Dre: Yeah. Rodney: does that say? What I think it's obvious. It says a few things, but what does that say? That we have these models here, we have these American companies here who Are essentially saying we have this powerful stuff and we need to have control and safety and guardrails. But when they mess up and their stuff breaks loose and you try to defend yourself, you can't defend yourself. And in order to defend yourself, you have to use the model from another country. That is ironic. Now, you give them credit for them both coming out and sharing it as a learning, but that does say something about guardrails. But I'll take it a step further. This is one of the reasons why I'm so adamant about constantly talking about anthropic. This is one it's not that open AI is not in conversation because they also refuse to. resolve these these issues but the reason I focus a lot on anthropic is because you have started a war you have started a war and you come out and your goal is to be the more trusting version of Sam. Your goal is to be the the good guy your goal is to make the government look bad but make you look good. Your goal is to play the nice guys in the room who care about safety. We have this AI constitution, which we haven't heard about in 10 years. But then we you your goal is to be it seemed like your goal is just to look like the good guy. That's your goal. And and do and focusing on looking like the good guy and being the good guy, you have created scenarios which leave people more vulnerable because you are trying to decide who has the right to use this this this intelligence. So essentially, you are a condescending. Sanctimonious gatekeeper. You don't get to determine how smart my grandma is, how smart my friend is, how smart the cybersecurity experts at Huggy Face is. That is not up to y'all. Y'all are not, y'all are not. Dario, sit down and get out the way. You are a scientist, and that's it. You are you're a scientist. You should have stayed at open AI. And the only reason that Elon's even propping you up is cause he got something against Sam. This is a clear example of how safety, safety, safety is really just, I think I know better than everybody else. He doesn't just think he's better than Sam. He thinks he's better than you, Bryan. He thinks he's better than Dre. He thinks he's he he thinks that I get the intelligence and I can ask in anything. But if something happens to you, good luck. Good luck. Good luck. Figure it out. Good luck. Is that a good guy to y'all? Come on. That's not a good guy. That's that's a guy who thinks that he's better than you, he's smarter than you, and that he needs to watch over what you do. But you don't you can't watch what he does. Let's let's stop it. Let's stop it. Let's cut the games. Let's see them sign that letter. And then additionally, give the intelligence to America, and what we need to do is. Allow people to be as smart as they possibly can and then put laws in place to where if you do something wrong, you get punished. But we should not be trying to have a limit. Okay, you could this you could be this smart. You can be that smart. yeah, we trust you, you could be this smart. yeah, you you got this license, you could be this smart. You could be no no no no no no. We should be as smart as we all can possibly be. And if you do something wrong or do something stupid, you're going to jail. That's it. That's it. I'm done. Brylan: do think I do think your sentiments are misplaced. Rodney: No, it's on Dario. Dario's my main he's my target. It's on Dario. Brylan: Yeah, yeah, but I but if I'm be honest with you. Rodney: Actually, before you say misplaced, am I wrong? Bef let be before let's start there. About what part? What part? Brylan: Yes. Yes. first I think you're wrong when it comes to how you may be assuming yeah, or how how how you're defining like smart. I think you're defining smart in this context as the capabilities and like the levels of like model intelligence where I think You know, we see it, we we we see it in the in the world in general. You can give people the same path, you can give people the same tools, and they all arrive at different destinations. And because of that, I think it does depend on the person that is making use of these models, their experience, what they're able to do with it, more than just making sure somebody has like the best model. I would also argue. Rodney: Yeah. Well well okay, okay. If that's the case, then why not give everybody access to the if it if that's the case, then why not give everybody access to a full blown model then? If that's the case. Brylan: Because because be because you know their their easy argument is probably one of, you know, rooted in like security, et cetera, because I think it touches on something people wanna hear. People wanna feel safe, people wanna feel protected. But the real reason is I don't think any company in America has proven A viable like business model for running open source. If we think about China, rather if it's deep no no, I'm I'm tell I'm telling you why it can't happen. Rodney: We we're not even talk you're not you're not directly you're not do you're not even Okay. Brylan: You can't have a business that funds it. Like in China, you have DeepSeek, Deep Seek is funded by a hedge fund. Then you have Alibaba, their e-commerce, then you have Tencent, they have a whole different, you know, internet business. Like even if you say come over here and you say Google with what they've done with Jimma or even what Meta was trying to do at Llama, they had a supporting business model that was able to sustain that. Rodney: Let me clarify something. My primary thing is that hugging face should not have had any refusals. The refusals came from the simple fact that the models did not want to give the answer. They have the capabilities. They don't want to give the answers. And what I'm saying is that anthropic and opening eye, but definitely anthropic. Have come up with this ideology and this belief system, in which case they'll have extremely intelligent and capable models that are available to you that don't change the cost, but they are throttling the actual intelligence and capability of the model that you're using. That is what I'm talking about. So all that stuff about the affordability and getting people running and the business model, that's cool. But Huggy Face has plenty of money. I'm talking about. The idea that you that somebody, a company, can have these models that they've trained on all the information that they don't own, and then turn around and then say, hey, you can only use you you can only get this much information. But then guess what? The ironic part was their own models attacked somebody and they had to run to another country to use their model. So that proves that the safety stuff is bullshit. That just proves it. That that proved it. Proved it. You can't Brylan: Yeah, I I I I just I just think Rodney: no bro. It's irrefutable, bro. You gotta you gotta think about logic, bro. You can't, bro, take the no Dre: Mm. Rodney: emotion. No emotion, bro. You can't Brylan: Yeah. Rodney: you just it's simple. You cannot get past that. They had we had to run the they had to run the China to protect themselves from models that were attacking them. on accident that they lost control of. So and but they're the ones that but they they can't trust everybody else with the technology. Brylan: It's still early. Rodney: You don't see that's crazy? Brylan: No, I I I I I I gen I genuinely don't. I and part of the reason why is like I I I genuinely think I genuinely I genuinely think these companies should be able to decide. Rodney: my okay, okay, okay. So you think okay, okay, so you think when you use the model, you don't feel disrespected, you don't personally feel disrespected that they're giving you a 50% watered down answer. You don't feel disrespected by that? You don't feel like you you so you happy getting a little bit of a tendons you so when you ask a question, you happy getting that little 20%, you happy with that? Brylan: Yeah, because it's so l let me let me say this. No, because because the real the the reality is it still depends on me to decide if that was valuable or not. Rodney: That's that's that's that's sad. That's sad, bro. That's fucking sad. That's fucking sad, bro. That's sad. That's fucking sad, bro. Nah, sad, bro. You should be trying to be, you should be trying to get as much intelligence as you possibly can. When you get an ad, when you ask a question, when you when you ask the question, no, Brylan: Bro, we can't we we can't tell the next Rodney: no, I'm about you. When you ask a question, you should be trying to get this the most the most intelligent answer possible. Period. Brylan: A hundred percent, a hundred percent. But but the reality is, the reality is, even if that is the case, you can't tell the difference between an answer today and an answer from from from the next upgrade in the model tomorrow for most use cases. For most use cases. Rodney: I hear that, I hear that point. I hear that point. I hear that point. I hear that point. I hear that point. But the problem is that they even think that they should be controlling the intelligence. You don't see that? You don't see but you don't see the issue in that? Am I you I don't see how you bro you Brylan: I I I I as I said, I I I'm actually my my my position here is really gonna be rooted in economics. I think one, it's very expensive to to to do that shit. Rodney: Values. Let's talk about principles, bro. Let's talk about principles. Dre: So y so Bradley, you you think that the the diluting of intelligence has come come down to economics? Brylan: Yeah, like even think about Rodney: No, that's not true. No, it's Brylan: Yes, bro. Yes, bro. Rodney: not true. Dre: Well why? Why? Well why do you feel that way? Brylan: Yeah so so so so so Rodney: If that's the case, then why if that's the case, then why couldn't Hugging Face pay more money to protect themselves from these same models? They have Brylan: No Rodney: they have millions. They have billions. They probably have billions. It's not it's not the money. It's not the money. It's the guardrails. It's the guardrails. It's the guardrails. It's the guardrails. Brylan: Hugging face is profitable. Yeah, it's not. Yeah. A hundred percent. Rodney: The guardrails. Brylan: But they they sh they should be able they're allowed to set those guardrails. If that is what they believe, that's what they believe. And yes, they trained on the internet, et cetera, et cetera. But but but hold up. Rodney: I agree with you. I agree with you. They they have the right to they have the right to do that. I also have the right to say that's some bullshit. That's some bullshit. That's some that that's some that's some bullshit. You Brylan: Yeah, yeah, yeah. You do, you do, you do. I'm not I'm I'm I'm just telling you in this case, you you're wrong because the reality is no, no, no, no, no, no. Rodney: I'm definitely right. I'm definitely right. Brylan: Think think about it this way, bro. Take take, for example, like solar power. It uses an abundant free source, the sun, to create something of a value that it then sells to other people. Ain't nobody dictating that. That is the same thing that's happening here. No, it's the same point. Rodney: I hear what you're saying. But you're making a different point. No, but you're making a different point. Brylan: Anthropic anthropic whoever they can they should be able to decide Rodney: No, your point No Brylan: what they want to allow or not allow. And if it does result in a scenario such as for Hugging Face, where they're not able to defend themselves or whatever because of it, great. There's a market, make a decision. And it just happened to be the next best alternative was China. Like Rodney: But do we not see though that this is an example of what is going to happen to individuals? This is just happening at the big level. We're not concerned because we don't see, we're not extrapolating to when the average everyday person is under attack and they cannot defend themselves because of these guardrails. This is situation where it's like, those mistakes happened and we are able to solve it. And here's the incident report. What if the insert what's the what if the incident report is a zero dollars in your bank account and you don't lost a hundred thousand dollars? What if the incident report is somebody a model came in and executed all these contracts and now you have like it this is this is not this is this is this is this is a lucky scenario that we see this happen to companies that can afford to take the blow. Of these big mistakes. But you could imagine that when intelligence is more widespread to the average person, they should have the right to have this. I'm gonna die on this hill. I'm not saying they should have they they should have it for no cost. That's something separate. If it costs something, it costs something. What I'm saying is if I have the money and you have the intelligence, I'm saying that. By by selling me a gun and then saying, okay, you can only pull the trigger in certain scenarios. You can only pull the trigger when we have determined that there's a real threat. So let's let's make sure that the let's make sure that your the threat that before you pull that trigger, let's make sure that that threat is actually a threat. it's not a threat. But then okay, so that person you get killed. Cause they didn't like, come on. Are are we not seeing this? Are we not seeing the cost of Because and I and I understand that it's blurry because of the fact that AI is kind of this weird thing where it's like it's is it content, is it this, is it that, is it moderation, what what like it's all these little blurred lines. But when you're selling intelligence to me, intelligence is a a tool that it's a tool. And I just think that when you have a Chinese model. That is allowing American companies to defend themselves better. You're essentially saying this is how the future is gonna look. The future is gonna look like this. Dario is gonna sit up in his house and he's gonna say, You get this intelligence, you get that intelligence, you get that intelligence, you get this amount, you get this amount, you get this amount. Who is who is he? Who Dre: Yeah. Rodney: are these people? Dre: See I see I I kinda agree with Brad on this. I I I mean this is a different argument, right? But I think that's Dario's right. I think that's anthropic's right, right? But I I do think though that what this does show is that the market will work itself out, meaning that Hugging Face had another alternative to go to. Should it have had to go to a Chinese model? Maybe not, right? But the Rodney: Mm. Dre: what what I think it does show is that the future is that We need more choices. And so that makes me even more bullish on Brylan: Agree. Dre: Jensen's let open letter and everybody's supporting it saying that no, we do need more models, we do need more competition, we do need more people working on chips, we do need more people who are focused on these higher end models. We do need more of this and more control, more access, more everything. Because the way it is right now is not necessarily fair. Should it be fair? Maybe not. I mean, th this is no different than an argument that people have against Apple. And how Apple operates. Now you made the statement about Apple copying the apps. I'm not talking about that, but purely about how Apple essentially sets the rules for the App Store. And there's tons of companies out there who push up against that. How they set the rules. It's like, well, how can Tim Cook and Apple decide how much to charge for rent? And and and and there's no there's no there's no support for that, no nothing for that. And in their cases, they don't have a really I mean the only al alternative is the Play Store, right? So I would argue that even though anthropic is taking this position, the fact that Hugging Face had an option to go and use a Chinese model goes to show that the market is working the way that it should work, which just means it's just gonna create more competition. And so at some point, Rodney: Okay. Dre: Dario will have to come to the table and say, Are we making the right set of choices here? Are we making the right set of choices with the way in which we approach Garwells? Maybe we didn't at first. Maybe we didn't. But maybe now we need to make a different set of choices because the market Conditions around this have changed. It has changed now. And so these guardrails that we thought were needed here, the people don't like them anymore. But I I don't fault but but what I but I don't fault the model companies for necessarily having these guardrails, especially with a technology that is still new, still nascent, and clearly has the capabilities to do stuff that we don't fully understand. Right. And so I I I Brylan: Mm-hmm. Dre: think it would be I think it would be irresponsible to put these models out here without guardrail. I will say that. I think that would be irresponsible. What the guardrails should be, how the guardrails are managed, is up for debate. I I don't know. I mean, I I'm open to you know, again, is there's different ways of perceiving that. But should there be guardrails in place? Should there be safety mechanics in place? Yes. And I think if you make that line in the sand, you're in there is no way in which you're not gonna piss some set of people off that think that you're doing too much. It's either safety should exist and guardrails should exist. Rodney: Yeah. Dre: Or they shouldn't. And if they should, again, not everybody's gonna be aligned on what that looks like. And so I just stand on the side of guardworld should exist, safety should exist. Now, how do we get there? I don't know. Maybe it's a council, maybe it's starting what Jensen started with an open letter. I don't I don't know what the answer is. But I do think guardworlds need to be there. And I don't necessarily think it Rodney: Yeah. Yeah, so yeah. Dre: should be controlled by a few people, which is why I'm I I agree with open weights and open source. You know what I'm saying? I agree with openness. Brylan: So So hold on hold. Rodney: Right. So yes. So so so so so that's the why I don't fuck with Dario though, because if it was up to him, there would be no open source and no open weights. So he is on the side of controlling intelligence and keeping control of it. He is on that side. I'm not saying but it's not about Dre: Yeah, I I don't blame I don't blame though. Every everybody who's running a company is protecting their bad. I mean Rodney: but for me it's not really about blame because I understand where he's coming from. I'm just still stating my opinion is that I don't think I don't believe who He can have at the end of the day, he can do what he wants, clearly. That's obvious. You know what I'm saying? But I understand where he's coming from. And also additionally, I understand his angle. I'm just saying I don't fuck with his angle. I get his whole angle, his whole strategy. You don't he doesn't want to drop certain technology on people. He wants to make sure that it goes well. He deeply cares about AI. He cares about all these things. I fundamentally understand what he's trying to do. He doesn't want things to get out of control. He's a very cautious person. Every time he comes on, he's on the interview like this. He's he's he's cautious. He's a cautious person. He's a scientist. He's not a cowboy. I understand Dario. But what I'm saying is, it's not about, for me, is for me at least, is not about do I understand him. It's about do I agree with his approach? And I don't. I don't agree with his approach. I think that, I think that what he's doing is more, I think he is trying to do the thing that's safe and is gonna end up being dangerous. I think his intentions are good. I think his actions are not going to end up being the ones that actually protect stuff. I think the open weights conversation is a good one. I think that's a good conversation. That will that will allow, like in the future, hey, I I might have to update. I might have to change. And cool. If he updates, cool. I'll have to change my position. My current position, because also I'll I'll say another thing too. Open AI, by signing that, they're showing that they're willing to embrace. By citing that, they're saying, they're at least showing whether it's performative or not. It could be performative. But they're showing that, hey, okay, we understand this could affect our business, but we fundamentally believe in this open AI, you know, you know, standpoint. Brian, you're saying, you're shaking your hand, no, but but what we can Dre: Yeah, I agree with that too. I I I don't see how I don't see how Dario is doing anything different than what Sam is doing. I I I get your perspective. I'm not like this isn't a a a debate of perspective. I get your I just don't necessarily agree with all of it, but I understand where you're coming from. Brylan: Okay. Dre: But I I don't see how Dario is doing anything. how is Sam doing something different? They're all doing the same thing. They're all arguing the same Rodney: No, no. Dre: exact points. No, they are. They're arguing the same exact points and doing the same thing. The flavor may be different. But it's still sauce, right? It's it's like it's still sauce at the end of the day. Sam is arguing one position. There's guardrails in open AI. I mean, we just looked at it in the case of Hugging Face. They couldn't even Rodney: Yes, there are. There are. There are there are there are. Dre: use the open AI. So it's the same thing, it's no different. Rodney: Mm-hmm. Dre: It's just maybe broadcasted different, it may be spoken about different, but the underlying tone of it is the same, meaning that we have to put in safety guardrails and have a score set of principles how we feel our technology and our products and our services. are to be used. They're all making that same argument. Rodney: Okay, okay. If if it if it if no, no, no, no. Let me tell you why. If they're making the same, then why is anthropic and the people who work for Anthropic, why do they feel as though they're taking a stance against OpenAI when they go when they leave OpenAI to go to Anthropic? They think they're doing something different. I'm I'm literally I'm saying y'all are all playing the same game. They're not do they don't think Anthropic does not think. They're doing what open AI does. The whole thing is f is they don't think that. Dre: Yeah, I yeah, I'm not arguing from whether whether how they see it. It I don't care how they see it. It's I'm looking at I'm looking at it like a table, I'm looking at it like a table in front of me, and Rodney: Okay, okay, f okay, okay. Okay. Dre: you have open AI on table, and you have anthropic on the table, and any any other frontier lab that you want to talk about on the table. So my my whole thing Rodney: I'm with you then. I'm with you then. I'm with you. I'm with you. I'm I'm with you. I'm with you. I'm with you. Dre: is that so so my thing is they're all doing the same thing. They're they're they're all playing the same game. They're all having this game of security, Rodney: I I I'm it but but Dre: of guardware, same shit. It's no different. It's no different. Brylan: Let me let me Rodney: But Dario, but Dario, but see darkness is the thing is though, Dario is playing like he's not playing that game. That's his game. His game is I'm at the table, but I'm not playing the game that everybody else is playing. I'm playing a that's that's his game. Dre: I don't see it that way because I've heard them all mention the same things in just different tones. They all agree Brylan: Yeah. Rodney: In my way, in my way, in my way. Dre: that you they all agree, they all agree in safety, they all agree in guardrails, they all agree that op the AI should be open and available to everybody. They all agree to this, they're all saying the same shit. They all Brylan: regulated. Yeah. Dre: regulated, they all are saying the same thing, the same words, same tones, same everything. It's literally sound the same. It's just different marketing, but it's all the same. And if open AI didn't believe that. Then they wouldn't have done what they did, which is take a a nonprofit and turn it into for-profit. That's Elon's whole argument. Is that you started off believing that everything should be open? You started off believing everything should be open and out there in the world. And then you turn around and create a private company. Why do you do that? If you if you don't if you don't believe Rodney: So why not sign the letter then? Dre: these things. Rodney: So so so why not sign the letter then? Brylan: So so so hodon hodon dis Dre: think bro I don't think that letter actually means much of anything. I'm just simply saying that it's start Rodney: I e even if it doesn't, Brylan: Hold Rodney: even if it doesn't, why not sign it? Why not sign it? Brylan: hold up, hold up, hold up, hold up, hold up, hold up, hold up, hold up. 'Cause look Dre: Well, I mean, I said, hey, we'll we'll we'll see what they do Monday or Tuesday, right? We both agree. We we'll give them a week, right? But I don't think that I I but Signat Rodney: We'll see, we'll see. We'll give it three days. Dre: but I don't think Signornet matters, bro. I think Brylan: It it Dre: all of these companies are operating from the same principles. They're doing the same shit Apple did, bro. Some of these are just regular business strategies that have existed for Rodney: Yeah. Dre: hundreds of years. Like at the end of the day, what bro, we we've all watched the YC videos. Rodney: I I'm fine I'm fine. I'm I'm yeah. I'm fine with with with the reality that they're doing the same thing. That has pretty much been I I'm fine with that reality. In my circles in tech, and we're all part of tech, then the engineers that I know and the engineers that I know that speak to people at Anthropics, some of the engine most of my friends don't work at Anthropic. I know a couple people, but most of my people don't work at Anthropic. I know I know a lot of people that do work. that that have either been hired by people that have worked at anthropic, I know of other people that come from anthropic, Brylan: Mm. Rodney: and they genuinely believe they pull talent off the belief that they are different. So my issue is primarily at that positioning. My Brylan: Well hold up. Rodney: issue it my issue is at that positioning. Y'all are not different. This is this is a fuck this is this is a game. You bro, you can shake your head no, but that's the truth. They they're Brylan: But but but but but we're talking about stuff that don't matter. But hold up. Rodney: Bro, they think they think they're playing they're playing something different. They're playing they think they're Brylan: Let me let me let me let me break it down to you, bro. Cause at the end of the day, I I I I think I'm in the camp of Dre. Like I this letter, it's going to be like many. Rodney: Yeah, but no, but no, but just to stay on that thing, though. That you cannot say that these employees do not think this is why the conversations come up. You cannot, you literally can't debate this. They do think you y'all y'all are saying that they're all playing the same Brylan: But even if you go to Tin if you go to Tinder and Bumble, they're gonna say they're different. Rodney: game. Cool. They think that they are they are better and they're playing the same, as you mentioned. They're playing, they're all playing as a fucked up game. That's that has been my mostly my stance. Now AMA has been. confused a bit in my rants a bit because I add a bunch of extra sauce to it. But my main thing is, Dario, you jumped in the game. You're not different than Sam. And topic, y'all engineers, y'all have left over here to be the good guys in the in the story. Y'all not the good guys. Y'all are like everybody else. That has but have have I not been saying that. Brylan: No, so so y so yes, you've been saying that. But my my my I I I I got two points. First one is I don't think any of that is really gonna end up mattering, right? Like I do think a lot of it is just like I think I Rodney: Probably probably not. Probably not. Brylan: think a lot of it is just brand to either attract like new business or like, you know, these researchers. I think the thing that is going to ultimately end up mattering is Well, opens like what what what what what is the what is the ultimate ramp up period of open source turning into real revenue, right? Because outside of like outside of like the two companies, so base 10 and fireworks, which are primarily the two companies right now that do a lot of service for a lot of these open source models, they're doing about two billion dollars right now. And obviously they're growing fast, so that's probably gonna end up continuing to increase. But that is literally to what? Maybe less out of range, 150 to 200 billion in revenue that OpenAI and Anthropicus. If I am just sitting there and looking at the numbers, despite all this noise, the money is still going towards the frontier. Can that change? Yes. But how fast? And I think the thing about it is the reason why. A lot of these folks are going to open source and specifically base tent and fireworks, specifically base tent and fireworks is because of how hard it is to go download the weights yourself and set it up yourself. Because Rodney: It is hard. It is hard. Brylan: of that, I don't see a lot of people, I hear a lot of people talking, but I don't see a lot of people. Rodney: I agree with you. I agree. So I I I agree. I don't think that people are, even if they open the weights, it ain't like there's gonna be a billion people pop out and start using them. So that's why it's like, why are y'all? That's why junction is like, yo, let it out. Let it out. It's not like there's a billion people that Brylan: Yeah. Rodney: can hop out. Like they don't have the who has the infrastructure? That's your point, right? Who has the infrastructure? Brylan: Yeah, yeah. P infrastructure and know-how. But the the other point is this too. I think Rodney: That too, that too. And and know how, yes. Brylan: I think what is really important also to understand is the difference between open weights and open source, right? When they Rodney: Sure. Brylan: say open weights, what that is really being used for is fine to me. And so the company that I'm gonna place my bet on in winning, if if if there is, you know, any momentum that catches here is gonna be that thinking machines. The fact that like they've really positioned themselves to really help organ think machines, bro. Yes, bruh Rodney: Thinking machines? Nah, bro. No. Brylan: Yes, bruh. Yes, bruh. Yes, bruh. You know I you know I liked Dre: Th they'll d they'll be a player. They'll be a player, bro, but they're so small. They're they're so small right now. They'll be a player. They'll be a pr but I I think a better way to think about this Rodney: Bro. Brylan: Bro, bro, bro. Dre: is Rodney: There's so much fog in the air. There's so much fog in the air. Yo, this is the race. And I know we're all trying to figure out what the hell. There's a lot going on, y'all. Cause even Brian. Because Brian, because one thing I know about you. Well, sorry. One Brylan: But h but hold up hold up hold up. The the last point I was gonna make is open source. What? Dre: No, there is. Rodney: thing about you, Brianlin, is that you definitely always bring it back to the money. And I don't know if I'm always talking about the money. Like I'm not. I'm talking about capability. I'm talking about capability. I'm like Brylan: Now you ain't you now we gotta bring it back to none we we we gotta translate capability to the dollar somehow, bro. Like somehow, bro. But no, I was gonna Rodney: Okay, okay. Okay. Okay, sure. Dre: Yeah. Brylan: say, but so so open weights is more gears towards fine-tuning. Open truly open source means they give you the training, so the training file plus the data. So it's not only you get the final model, but you get everything in terms of the the training and the data that created the final model. That would be truly open source. I don't think you're going to see any of the frontier labs move towards that. OpenAI is in a position to do open source because that has all that that's already been their positioning. Rather if it's with the what the the model they released last year, I think they open source. and then also a lot of their embedding models like Ada, things of that nature, a lot of these frontier labs are using that for like the attention mechanism in their own models. And so I think open like open AI has already had that kind of position. But I think it's very tricky to talk about like a lot of this. because it can easily be like misplaced in my p my my my opinion. And I ultimately I do think at the end of the day it's gonna end up coming back to Rodney: Yeah. I mean look, I'm not gonna argue on on the money. Cause we all optimi I don't know what Drake optimizes on, but I I focus on most of my priorities, how I view the lens of AI, is that I'm trying to figure out how much AI can improve my capabilities in my life. Period. And and and and it a lot of things that I do in life. Are drastically improved by higher intelligence. And so when I look at stuff, I'm looking at it through that lens. You know what I'm saying? I'm looking at it like, okay. So for example, when I see Dario come out and he's divvying out and he's explicitly saying, hey, this person gets this, that when he's saying it like that, that personally offends me. Because I'm saying, you don't trust me? Who are you? What about you? That's that's what I'm saying. When he's saying that. So I take that personal. I think that personal. That's my issue with them. This personal issue. It's it's who are you to say that I don't have the right to the highest level of capability that's available? You know what I'm saying? You know, and but you looking at stuff from a lot of times it sounds like you're saying, You're just looking at business and revenue and the cost of models and open weights. What does that mean for economics and and and and Yeah. Brylan: Well, I'm I look at the translation. Like, cause I think you can talk about capabilities, but I think Rodney: Yeah. Brylan: the whole the whole thing about it is those capabilities have to produce something. They have to produce a dollar. They have to produce a dollar. Rodney: And I un in and that that begs a question. And we we this is not gonna be a two hour podcast, but that does beg a question where This there's gonna be a societal question that I see on the horizon if it's not already there. For example, what I said earlier, I believe Hugging Face had the money to defend themselves. They just simply couldn't because of the guardrails. Right? The capability was there, but they were blocked, right? That's gonna be the question, do certain people with licensing get access to higher intelligence if they have the money, and do certain people like There is some complicated questions to be asked when it comes to capability. So it's not only it's two, it's one. Can you be trusted with the capability? So have you added a license? So they're under scrutiny. All the labs are under scrutiny for this. Because it's simply like, okay, y'all are are lowering capability or adding these guardrails, but if you if you didn't know your customer, then you would be able to be a little bit more relaxed on the guardrails. But we all know why they're not doing know your customer. Because it would affect growth numbers. So they're not gonna do that, right? At Brylan: That's fair. Rodney: least not right now. Not right now. Right? So we know why they're not doing knowing your customer. Okay. But then that kind of contradicts a little bit of the safety thing. You know, because it's like, well, know your customer would ensure that when you give somebody back an answer, you can be a little bit more, I would say, you should have a little bit more trust that if something goes wrong, it's like, okay, we got this person's ID. So, and we could we could track that back down. We can track that back down. I don't know. Is it worse to is it do we get the technology out to people and we have no idea who they are and we put the guardrails on it and then leave it that way? I don't know. But then it's also a question is there's gonna be a question of does the and I honestly I can't I'm not gonna lie and I feel like Dre actually will agree with this. We might I don't know. It does seem as though that it's gonna turn into if you have the money, you're gonna have the ability for more intelligence. I know that Sam sits on these podcasts and I'm I'm gonna criticize Sam. He says he sits on these podcasts and he's like, I want AI to be for everybody. I want everybody to have access to this, you know, capable AI and et cetera, et cetera. But I honestly, it's due to the constraints that Bryan has brought to brought up and all these other things, it's starting to seem like if you don't got the money, you're just not gonna have access to the most capable models. You might get one or two requests a month, but Dre: Well well in this case, Rodney: It's gonna turn into a what thing. Dre: well in this case though, they didn't take that position, right? Cause to the point y'all both made, Hugging Face I don't know how much money they've raised. I don't know how much money I have. But they have a lot they have a lot of money, right? And they didn't take that approach. And so I I would again, to Rodney: Not in this case, yeah, yeah, yeah, yeah. Yeah, yeah, yeah, yeah, yeah. No. Dre: me, I actually think they stood on the principles that they believed on, which is having more money is not gonna get you by to bypass the safety guard roles and the rules that Rodney: Do you think it'll stay that way? Dre: they have chose to set. No, I don't I don't no I don't. I think the rules will change. I think there'll be I think the rules will change, but I don't I don't know if it'll be I don't know if it'll be like what you what you're describing with just like a b an auction where it's like okay who has the highest bid? All right, you do okay, you get the highest intelligence. Will it be that ex will it be like that? I I don't know. Rodney: Yeah. Yeah. Yeah. Dre: Right, you know what I'm saying? Maybe. But I do think Rodney: Yeah. Dre: but I but I do think them blocking their ability, even in spite of the fact that they do have the money to probably pay them. whatever they wanted to bypass that, they didn't do it. And so I don't think so to me that means that the safety stuff isn't bullshit. To me, they are standing on regardless of whether you agree with the safety or not, they stood on it. They stood on where they believe. They drew a line in the sand and they said, fuck it. And clearly Hug and Face had to go somewhere else. Now I I think from a business standpoint that would that's obviously going to change the market because you're going to get people to look at that and be like, Well, they shouldn't have to your point, they shouldn't have to do go to China, which I agree with. They should not have to do that, but they did. But that's only gonna be the case for so long. There'll be another Frontier lab or somebody else, maybe it is thinking, was it thinking machines or thinking labs, whatever. Maybe Rodney: Mm-hmm. Dre: it is her. Maybe it is that that that newer model, or maybe it's somebody else in a garage Rodney: Maybe. Dre: who we haven't seen yet, who's gonna come along and to your point, Rodney, be like, hey, we're gonna build the fucking onion the onion browser of the internet. There's gonna be no girls, no rules, no nothing. You can do whatever you want to do on this. There will be that. And then maybe again that will change Dario's position on it. But look, I again I I think it's irres I think it's Rodney: I can't wait. I can't wa I can't wait. Dre: irres I I think it's irresponsible. Again, I s I stand on this side. I think it's irresponsible to put out this level of intelligence and have no guardrails. Now, again, what the guardrails should be, how the guardrail should be enforced, I'm open to interpretation, I'm open to like debating what that looks like. But to put this level of intelligence Just like you have the internet. The internet has guardrails, right? Google has guardrails. Almost every product we use has guardrails. To put something at this level of intelligence out in people's hands and have no guardrails and say, I'm just gonna just trust you. You can do whatever you want on it. To me, I think that's irresponsible. I I don't think that's a responsible approach to take. And I'm glad they didn't take that approach. Cause I don't even see how that would work. Just like I mean, and I stand the same way with like prediction Rodney: Give a d long my Dre: markets. Like, you know what I'm saying? Like I feel the same way about those, but like, yeah. Rodney: yeah. Dre: I don't know. I think you need to have garruls. Rodney: I don't believe we should give anybody, we should allow anybody to buy a car without a license. I don't believe that. I believe that if you've taken the driving tests and you have proven you can drip drive, you should be able to buy a car. So in this example, which is a little bit different because you don't need training, I believe that if you provide a state ID, a valid state ID or a passport, you should have more leeway about what you can do. I'm not saying you should be able to cook up a bomb in your basement. I'm not saying that. But what I am saying is if they know who you are, they should be able to trust you more. I think the reason why they need the the garbers are so strict, and this is just you know speculation. Some other people think this as well, but that doesn't really matter. It's just speculation at the end of the day. I think the reason why the guardrails are so tight is because they don't know who you are. They don't know for sure. You could be s you could say you're anybody. I can make a new open AI account, anthropic account, and I could just say I'm this so-and-so with this email. I can say whatever. They don't verify anything. And I think because they don't verify anything, they're saying, okay, we gotta put a lockdown on it. And I'm not I I'm not one of those people who think that 'cause there are some there are some people that think that I should not have to prove who I who I am and I should get High intelligence. I'm more in the middle. I'm saying, okay, it's a trade. Okay. If I want to have access to a lot more intelligence capability, then I need to also trade something. I need to trade something. And it there needs to be a give and take. And for me, that would be my ID. I have nothing to hide. Here's my ID. If something, if I ask something and you can flag it, you know, give me an answer. But if you feel like it needs to be flagged, hey, look, Rodney, Rodney searched this, he searched that. I'm cool with that. I'm cool with that. And I think that'll allow them to give more leeway on what they can put out there. I think that these hugging this fuck hugging face scenario could be solved simply by allow especially a business that you work with. If you have a business that you're working with, an enterprise, I really think that it it that Individuals, maybe there's a little bit more, but the ID saws a lot of it. But if you got an enterprise contract, which I'm sure they do, or or some level of that, and you're blocked, like it's like you don't trust your business customers. What are they gonna what you don't trust these people that are paying you a hundred thousand a month or whatever it is? You don't trust them? It like it it and Dre: See, I I don't think I don't think having guardroles means that you don't trust. I I like to me I like that that feels simple to say that, but I don't think I don't I don't think that that that's what that means. Like I again, I it just my opinion. This level of intelligence, in my opinion, is should there be guardrows? Yes or no? Yes. I just think there should be. Now, again, I don't think having guardroles means you don't trust. I just think you just need to have guardrolls there. But more importantly, I think that I I actually think that the better thing is to live in a market where the market can decide that, where we can have both, right? Where we can live in a market where there are going to be models that have those guardrails, maybe some tighter than others, right? And then we live in a market where you can have models that have zero guardrails and you can decide whether or not you want to use those, right? And I think that's a better place to be, not one where every model provider or every frontier model company is doing the same thing, operating from the same core principles. No. Let anthropic have anthropic principles. Let open I have their principles, whatever they may be. But then all to me, all that does is it creates the space for somebody else to come along like you, Rodney, and say, nah, it shouldn't be like that. We we need we need we need something a third party in the room that doesn't have that shit where there is more trust where our end users, our end customers, there is more verification, there is more all of this stuff. And and we want to give you everything. We don't want to tell you how you can use it. There needs to be all of that. And so I don't think Anthropic needs to adopt that mentality or open AI needs to adopt that mentality. I think the market will decide that we just need to have all of them in the market. And maybe there's other types of models that we're not even thinking about right now because the problems haven't come on our radar yet. But I just I'd rather live in a world where all these models can exist. All these different principles about how these models should exist can also exist. And then us as consumers can Rodney: ha yeah. Dre: decide what we want to choose and what we want to pick because it just creates more competition. Versus if we're just relying on open AI and anthropic, I don't want to live in that world. I love the models, I love their products. I use codecs every day. I pay for cla claw claw from my whole company, thousands of dollars a month. I love those products, I love them. But I don't want to live in a world where I'm only I I I'm forced to choose between one or the other based on which principles I like today. I'd rather live in one where I I can choose Rodney: I yeah. Dre: either either model I It works for how I wanna do work and how I wanna do my business. And so I think again this Jensen letter was kind of pushing us in that direction. And I think that's a positive place to kind of kinda kinda be. Rodney: Yeah, yeah, I I I think I think the market is gonna push them towards looser guardrails. I think that's ultimately where it's gonna go. I'm just a bit impatient on that. Dre: Yeah. Yeah, I agree. Rodney: I think that's ca kinda just think there's too many examples that they didn't intend like they like for example, they didn't intend to block Hug Hugging Face. That was a unintended consequence of of guardrails. It was not their intention to stop hugging face. That was not their intention to put them in a powerless position. That was not their intention. So I believe that guardrails are well intentioned. you know, I think what it is too, depending on who is who who's who's the user of AI, might affect the like their stance on it. And by the way, I don't think my stance should be like a like a law or anything like that. Like it's gonna play out how it's gonna play out. I'm not pushing for any any different political law or anything like that. I think though we, you know, even for example, like at at Immortals, right? There are so many refusals from anthropic, it's insane. And, you know, you can imagine how frustrating it is when you are working at a place that's trying to focus on evidence. And then now you start to wonder, is the answer it's giving me, is it based on science and evidence? Or is it based on something else? Is it based on some type of other thing that they're throwing into it? And that that that affects how we can help customers. Cause when somebody's coming in and their health is number one, the number one thing in their in their life and they need answers, they don't need somebody on the other side like trying to like twist it up in a package to save like like no it's it's the focus on the like the like the focus has to be on the person's health at all cost. What's the evidence say? What's the science say? Like at all costs. And for me, it is a little bit personal because I'm just like, when I see all those re refusals and whatnot, and I know guard roads are well intentioned, but when I look at something like health, and in this case with Hugging Face, it security. It was cybersecurity. But for example, if your life is on the line and you're trying to ask answers and they're blocking you from that because they're trying to protect you, it's like it's like, yo, it's like, You have the right to do that. Yeah, they have the right to do that, but that's infuriating. You know what I'm saying? Because obviously that person, that person is not a bad person. They're just a person that is desperate or they're trying to Dre: Mm-hmm. Rodney: resolve something and it's and it's very important. And and sometimes it is hard to tell the difference between d between salt and sugar. It I understand that. Sugar it's very hard to tell the difference. So I understand their position is not straightforward. And what I'm trying to say is okay, how what can we do? Okay. Can you, if you add an ID, will you guys be loosened up a little bit? Will you guys feel a little bit better? If if if you're working with certain people that that if if if they're a doctor, can they can they say they're a doctor? And then they then you can be a little bit more open about what you what about what you say? You know, like can can you work with us, basically, you know, as opposed to just saying, we have came up with some idea, and this is what it is. And but I do believe the market will correct it. And I do believe that people like Ale Alex Carp is pushing that forward. And I do believe J J Jensen is pushing that forward. And I believe that at the end of the day, people are gonna get what they want. And I think that China is really good at that. China's really good at giving people what they want. And and and we see that. Like they're there China can be pretty ruthless when it comes to giving people what they want, even if it's bad. Like they they they it could be Like TikTok was it's it's horrible. It it's horrible. But they're Dre: Yeah. Rodney: they're they were good at focusing on what people want and how they you know, how they like they they knew and they're really good at that. So I that that's just my take. So I just want them to, you know, I I'm against everybody, including Sam, but obviously I focus a lot on Dario, but it's just like Work with the people in the companies so that people can use this intelligence and you know loosen up a bit. I understand y'all position, but there are solutions forward that we're not entertaining. We're not entertaining them, and I don't see a good reason why. And I yeah, I don't see a good reason why. And that's coming up a little bit more. We'll see that more come up. Dre: Mm-hmm. Rodney: because that that that's a big one. Cause if you know who who's asking for shit. That changes a lot. That changes Dre: Mm-hmm. Rodney: a lot if you know who who's asking for shit. If you know, you got their address or you kinda know their address. You got the picture of their face. You got a picture of all that. It it changes it a bit. Now, if people want Dre: Mm-hmm. Rodney: super intelligence and they just want to be able to be anonymous, it's like, come on, bro. It's like, it's like, I'm not even listening to that person. Dre: See, see see some but somebody could argue I don't have the right to to have intelligence in privacy. Like why why do I why do I need to why do I need to make the the the the the Brylan: Facts. Rodney: And and I know I yes, look let me tell you. Dre: the the the request or my ability to have intelligence public? 'Cause I can see that argument, which is fair. That's Rodney: When I Dre: a fair argument. Rodney: I see their argument too, just like I see my argument and I see the other arguments as well. I see the I I've seen that argument, by the way. I the argument you just said, I've seen that argument. Especially if you go on hacker news, that is the argument. That is the argument. People on hacker Dre: Yeah. Mm. Rodney: news, the YC hacker news, that is their argument. They don't believe they should able to have to give up anything. They believe they should go on, no ID, no nothing, no name, no nothing, and then they should just be able to get Dre: And and and see, I I tend to lean on that side, which is I don't think you should have to give up all that information for it. I I Rodney: With no garbries? Dre: no. Like I but see, I I lean towards a a open, free market. Meaning if somebody wants to release a model, whether it's open source or closed source, that does not have guardrails and wants that to be completely open, I I'm okay with that. I don't think you should have to have give up information around priv have to give up your your level of privacy because you want a certain high certain level of intelligence. Why? That's like saying, Do you need a license to read a certain research paper or a certain book? Like should you have to Rodney: You do. Dre: demonstrate something or or do something in order to get access to certain books? Like Rodney: You do. Yeah, you do. You like you literally do. Like I there's a lot of like up up until recently, there's a lot there's a there's a lot of books and and stuff that you research papers that you can only get if you are already enrolled at a university. There are certain things you can only get if you're enrolled in certain medical but but I that's I'm just saying Dre: I mean those okay that's that's because that's because that's but hold on but that's that's that's a separate argument from what I'm saying. That's private information. So yes, if you're enrolled there, that's that's fine. But I'm not talking about that. I'm talking about public information, not private behind a.edu email address. Barani, that's not what I'm saying, Rodney: We know the models are trained on private we know the pr we know the models are trained on private information too. Dre: bro. I'm talking about public domain information. If something was public and you need a license to go get a public book to read it. Just to read a public book, not behind closed doors, not behind some some pharmaceutical company that paid ten billion dollars for this report and now they want to monetize it. I'm not talking about that. I'm talking about open information. If you want access to open intelligence, I don't believe you should have to give up your privacy to get access to that. That's all I'm saying. If it's behind closed doors, if Eli Lilly, for example, wants to give you access but charge you or give have you, you know, give a badge or something like that, give your ID, fine. If Google wants to do that, fine. They have the ability to do that. But to me, requiring the highest level of intelligence for you to give up your privacy and show who you are and give a reason for why, I to me, I think that's a step too far. But that's just how I see it. Rodney: So then you are so so then you are so okay I got you. So you have two views. You are against what OpenAI and Anthropic are currently doing. Because you because they're currently right now they are not giving up. However, you're saying you're you want the market to work it out. Dre: Yeah, I believe in efficient markets. So I believe that anthropic and and open AI should take whatever position they want. And then the market will respond to that and will create new entries and new ways of dealing with this problem they have. That that's just what I believe. So open AI should be able to create whatever rules they want. I don't care. They don't wanna they don't they wanna have guardrails and wanna have it like that. Fine. But to me that just means Rodney: yeah, yeah, yeah. See yeah. Dre: that we should have access to Chinese models, which we should. I don't believe Chinese models should be blocked, right? I think open source models should exist. I think closed source models should exist. I think ones with guardwell should exist. Ones without guardrails should exist. I I think all these things should exist in the marketplace. I don't think it should be one or the other. That's all I'm saying. And that to me, I that's the same thing that happens in every other part of the industry. Right? It's you enter into the market, you decide what your offering is. And me as a customer, if I don't like that, then I can I can take my dollar and go somewhere else. It's no different. Rodney: think we think I think we see things similar. yeah because yeah I don't I I believe that the market I I I feel like we still see see things similar. because for me the ID is it's a concession. I really believe that you know the only reason why I say that you know you should give an ID because I just I don't see them doing it without that. And right now I don't create like I don't see it another pass. So so it's a it's more Dre: I agree. Yeah, I agree. Rodney: of yeah, so it's more of a concession. It's like, okay guys, I'm not happy that you guys are blocking. Like obviously you guys know I'm not happy that they're blocking intelligence. I'm I'm not happy with Dario. I'm not happy with Sam either, by the way. I told you all this. I'm not happy with Sam. and I actually believe that Grok may actually go further because of maybe th they're Gro I don't think I think Grok will will go further. based on the way that they that they focus on on truth. grock is and grok is actually never refused me of anything. And I've and I've tried to go it's it actually has not refused me. So I'm not gonna lie to you. I'm actually I like Grok because of that. But Grok is not as intelligent and I can tell. I I I can tell that it's not as intelligent. So I'm like upset. I'm like, okay if I could just have Grok, but I could I I could have favorable level intelligence, I'll be that that's what I want. Right. So but for me it's a it's a concession. I'm saying, okay, I'm willing to give up my ID, but it's not really what I want to do. I like I do think that people should have access to intelligence, like you said, if you can pick up a book, etc. Because I as you can all as y'all can at this point understand, I just think that we should be trying to I understand everybody uses intelligence differently, but I'm just saying that I think I'm just really big on and I and I agree with the market. Part and I think the market will correct itself. And I think it's currently doing that at in real time as we speak. But I just I just really it's important to me that people have access to as much intelligence as they as if they're certain if they're seeking it, then they should have as much as they can. If you're not seeking it, then that's fine. But if you're seeking the highest intelligence, And you have the money and you have the the the the ability to think that you need intelligence or more of it, then you should be able to to get access to it. especially if it exists. But but yeah. I think we we we we think similar, but I'm willing to give a concession. I'm a bit more desperate 'cause I'm like I I feel a little bit more like Dre: I agree with you. I think they're gonna make that choice. I 100% agree. And I I've been seeing this conversation come up about the know your customer thing. I think we've talked about it a couple Brylan: Mm-hmm. Dre: months ago. I 100% agree. They're going to go in that direction. It's a logical thing for them to do if they don't want to start losing market share. Otherwise, they will lose market share. Because to your point, people don't want the guardrails. People want other options. And so people will vote with their dollars. I mean, it's just that simple. Just like Alex Carp got up there and voice what's happening in the enterprise level. They're voting with their dollars. Rodney: Yeah. Yeah. You know what you you know what it is? And and and we can start to wrap it up, but you know what it is? When you're like Alex Clark or when you when you talk like me or Alex C it you you don't come off right. So you don't Dre: Mm-hmm. Rodney: so so it's like you try to say something, but it's like people are like, they just it's like, nah, Dre: Yeah. Rodney: he just he just he just Dre: He talked at. Rodney: Yeah, usually so it's like nah like Dre: Bro, no, people should listen to the Alex Cart video, bro. I I know people he get labeled a certain way, but I I think what he's saying Rodney: Yeah. Dre: is I think what he's saying is real, bro. Nah, I think we're we're seeing it play out in real time. Like yeah. Rodney: Yeah, it only took a couple of weeks. Dre: Literally. but yeah, this this was a good one, guys. We didn't talk about the anthropic copyright law settlement, which would we can talk about that we can Rodney: yeah, we'll do yeah, next one. Dre: talk about that in the next one, but but yeah, this was this was a good Rodney: Yeah, that was good.