SPEAKER_04 0:00

All right. Hey Dustin, how are you? Hello, Hamid. I am good.

SPEAKER_01 0:04

How are you? It's been a hectic week, man. Holy macro. Yeah? So much has been happening, and I've been traveling on top of it. So it's yeah, it like literally just uh came into town, went directly to the office for the podcast recording. So here I am. Glad you were able to make it.

SPEAKER_04 0:24

What do so okay. For this week, if you couldn't, if you if you aren't on YouTube or you're not watching the the feed, we have a new bar on the bottom to help everyone follow along on the agenda for for the episode. So and to help me follow along. So we're gonna talk about that's a C bar. I like I like that. That's yeah, it needs it needs some tweaks, but we're going to improve it uh next episode. It'll just keep uh getting better and better. Um, so we're gonna talk about all this AI beer. I know we kind of talked about it, I think it was two episodes ago, but it's come up again and again. So we'll we'll give our thoughts. Um, there's a lot of drama around Robinhood and its stock tokens. We'll break that down. MetaMuse has been absolutely killing it lately. So we'll talk about that. There's some bumble news, and of course, as always, listener questions. So let's start with Micron and all this AI fear in the market. Um, so I think to start, Micron was down like six percent on Monday, if I remember correctly, from from the this fear. Do you do you think this is justified? And what are your overall thoughts on what's going on?

SPEAKER_01 1:54

So before before we talk about Micron specifically, it might be good to talk about what is the AI sort of saga that is going on. And it sort of started with um this anthropic engineer that left anthropic and uh said in a post on X that he worked for both OpenAI and Anthropic, and neither company is uh serious enough about the security threats to humanity, basically, of AI. And his post on X went super viral, largely because there's a lot of people who are working on AI who share the same fears.

SPEAKER_04 2:30

And um and sorry, to be clear, didn't was it him that said, or was it a different tweet that said there was like a 10% chance of AI wiping out um humanity in like 10 years?

SPEAKER_01 2:40

Yeah, it was him, and it was that's that was like sort of his guesstimate as to you know the 10% uh chance, which is obviously a made-up number. There's no way to actually know any kind of probability of something like that. Um, but he did say that like there's a lot of people in his um sphere, the people who are working on AI that are concerned about this and concerned about what it might mean to the end of humanity or or humans, as especially once it hits um recursive superintelligence, meaning like the superintelligence that's improving itself. Um and um and basically uh that tweet went so viral that the uh the camp that is on the other side of this uh this uh position, which is that we need to go full speed ahead with AI, called it a PsyOP. This is an a plant. They sort of planned this whole thing. This is not natural sort of uh uh virality of a tweet. Elon sort of doubled down on it, saying he's never seen a post like that go so so viral so quickly, especially from a relatively new account. Um, so basically, everybody on the other side of this uh AI uh discussion was that this is this is bologna and this is sort of a plant and it's not real. And that's not really what uh the people are thinking. Then Dario came out at in interviews and and uh said basically he agrees with the person who left, and he uh posted a proposal, a three-step or three-phase proposal as to what to do about it, and that we need to sort of slow down um the pace of the frontier models. Um and Dario sort of like additionally scared a lot of people, and his interviews sort of also went viral on the media. And you can see why the media would be eating this up because this is you know an existential fear that everybody has, nobody wants to go extinct. Uh, and Dario sort of like fed fed into it. But but then suddenly Elon changed his position that, like, okay, Dario is right, and we probably should uh have some oversight on these frontier models. And by the way, I have been warning about these um concerns for more than a decade. So um so basically uh the fears were realized by essentially all the sort of major players that matter. Then uh the all-in summit happened uh starting Monday. Uh, and of course, Jensen uh Jensen Wong, CEO of NVIDIA, sort of weighed in that this is irresponsible, uh, what Dario is saying and what other uh people are doing, creating fear uh for people who don't know what's going on. During Jensen's discussion with the uh all-in hosts, uh President Trump calls into Dario, goes on speakerphone, and he also says, this is a hoax. This is just like like we need to go full speed ahead on AI, we need to beat China. Uh and of course, the all-in summit people uh host themselves, that includes the AI czar for the United States, uh David Sachs, he also views it as a hoax and that um it's basically a Chinese ploy to slow us down or the democratic ploy to slow us down. Some some Democrats have come out totally against AI, um like like Bernie Sanders, for example. So um so the the sort of like sentiment in mont on Monday morning was that uh no, we need to go full speed ahead, and they're they're basically wrong. Then in the afternoon, uh they had uh Elon Musk as a surprise guest and they talked to him about you know these AI fears. And the way in which Elon phrased it is that if the people creating the frontier models are telling you that they are afraid of the stuff that they're creating, you should probably listen. Uh they're not playing some 4D chess, trying to like uh tell you that uh our product is gonna kill you in the morning and in the afternoon ask you how much stock would you like to buy in our company, right? So um their fears are probably real. And this is something that is has been prevalent. I mean, like I've seen TED Talks of people who have been the godfathers of AI who've been talking about these fears. There's um co-founders of uh uh OpenAI, uh various different people in both OpenAI and Anthropic and Google who have left over these concerns. So the fear within the people who are creating these technologies is absolutely uh real. And Elon sort of came up with this uh endorsement of phase one of uh Dario's plan, which is essentially to plant people into each other's companies that might check each other's homework rather than testing your own um model uh for uh the companies to sort of test each other's models before they're released. And this plan seemed to be something that um uh at least Dario and Sam Altman kind of agreed with. Uh, and then uh David Sachs, sort of like on the fly, sort of agreed with Elon Musk and kind of changed his position, as well as some of the other uh all-in summit hosts. So by the end of Monday, it did seem like suddenly we were in this uh uh this sort of agreement that maybe we should be checking each other's homework on the frontier models specifically before we release these AI models. Uh then yesterday, Mark Zuckerberg tweeted or posted on X that listen, uh AI security and safety is everybody's responsibility. Nobody should be releasing a product that doesn't do what the customer wants. And uh, for example, we delayed the release of uh Meta Muse for months to make sure that we got the safety and privacy issues right and uh uh the aligning the model correctly before we released it. And we didn't you know go out there and tell everybody else to pause their development in order to do it. And suddenly his position seemed like the the sensible one to take. And everyone sort of like jumped on on board with uh with Meta's position on this. So that is the background that uh uh of the AI discussion that we're talking about. Now, there's a lot of misunderstandings and misconceptions here. Some people have taken Dario's views or uh the AI community who wants to sort of like slow things down as we can't stop AI, nobody should be stopping AI, we can't stop open source, we can't stop China, and therefore, like they're all dumb and this is stupid. Well, no one's asking to stop. They're just saying slow down and possibly go through some more checks and balances. Like if there's a 10% probability or whatever, some non-zero chance of uh you know um AI potentially uh killing humanity, then we could actually take paths to make sure that um that we we increase the probability of a better outcome, right? More aligned outcome. So that's not what they're saying exactly. But let me pause there and then have you, you know, uh ask any questions or clear up any misinformation that I provided there.

SPEAKER_04 9:47

I mean, no, I think that was a good overview. Most of bringing this back to Micron, and I have my own opinions, which I'll I'll bring in in a second, but the sell-off with Micron really happened initially um when Dario, I think he he posted his tweet. Um, or sorry, no, it was um the employee posting the his tweet, and maybe Dario had a response at that point, I don't remember. But a lot of like the Elon stuff, the Mark Zuckerberg stuff, like that didn't happen yet. And Micron's been fairly stable since Monday.

SPEAKER_01 10:24

Yeah. So Micron uh uh went down, I think Monday Monday itself uh it went down like six percent initially. Um also Nvidia went down uh some on Monday. So the the AI trade or the semiconductor trade did get hit somewhat. I think that uh from a from a stock standpoint, it seems like those fears that the that any kind of slowdown might mean that these companies are not going to hit their revenue targets. I think that fear is sort of like uh not um doesn't seem justifiable because uh, first of all, the demand for their products across the board, whether it's Nvidia or uh or Micron or some of the other uh semiconductor companies is so high that even if we slow down pretty dramatically on frontier models, that doesn't mean that the demand for these chips are going to be any less, right? Because the usage of the product oftentimes is actually coming from um employing these models in lots of different areas. So today or yesterday, for example, um uh, or maybe it was today, actually, like all the days are now meshing together, but uh anthropic uh and Salesforce sort of join forces to um to integrate uh Anthropic's product into Salesforce better. So like there's a whole new use case and uh set of um AI needs that are going to happen with an existing model that's getting integrated better into Salesforce, right? So a lot of the um fears of what a potential slowdown might mean for an Nvidia or for a micron seems unfounded. Um, but you know, the market is the market. It sort of likes to react to the smallest things. And here we are, sort of like having it react to uh the discussion of how we should be responsibly building these things going forward.

SPEAKER_04 12:12

I mean, the death of humanity is not a small thing, Humid. Um okay, let me let me give my perspective on this, which I think ended up being similar to what Mark kind of wrote. Um so as you know, I'm probably one of the greatest product managers uh alive today. Ever. Yes.

SPEAKER_05 12:33

Ever, yes.

SPEAKER_04 12:34

And over time I've developed uh several guiding principles of how I work on products. Now I don't work on AI products or like AI models, but the the principles do work kind of for anything. So I have this principle called the Jafar principle, which will make sense in a second.

SPEAKER_01 12:57

I assume it comes from Aladdin's Jafar. It does. Okay, this should be very entertaining. Go for it.

SPEAKER_04 13:03

The the principle is if I'm developing an extremely powerful feature, but it is not usable, usable in the sense of it's too difficult or cumbersome, even because it's such a uh extreme action in the product, right? Like it's difficult to make those uh user experiences simple. So if you have this really powerful functionality that's difficult to use, it's not worth adding that that functionality at all. Because if you can't use it, what's the point? And this is very similar to where we are with AI. And to bring it back to Jafar, at the end of Aladdin, you have Jafar, the villain of the movie. He wants to become the most powerful creature uh on the planet, which is a genie. So he wishes to become a genie. What he forgets is that even though you have all the power in the world, you cannot use it unless if someone wishes you to do so. So he has all the power, but he cannot use it. And if you have these frontier AI labs working on new models that are uncontrollable, what's the point? There is no point. That's not that's not a viable product. What they should be doing, if this is a real risk, they need to take the time to make sure whatever the next iteration of these models are can be fully controlled by the prompts that users are entering. Now, when the result of that might seem like a slowdown, but I don't frame it that way. I frame it as it is how you need to innovate to get to the next level or the next frontier. Right.

SPEAKER_01 14:54

It's so the argument. The argument there is like, are the car companies slowing down and collaborating to slow down to make sure that they're building their cars safely? No, they're not. But but like I get that. Like, and I think that there's there's multiple concerns with AI safety, right? Like, there's the the safety that um uh people will use AI in an inappropriate way to do something nefarious, like poison the water system or create some sort of um uh virus that infects a lot of people and kills them or whatever. And that's similar to like people using guns to kill people or you know, governments using weapons to kill you know many thousands or millions of people. So that type of uh that type of use of AI, ironically, will probably continue. And we won't be able to stop it no matter what kind of regulations we put on it, because the governments are above the law, right? Like they the governments themselves do not actually adhere to any kind of uh these type of regulations. They're like, no, we want a version of the the AI that's gonna do exactly what we want it to do. Uh, and that has already been fought between, you know, like the military uh contract that uh that they they had with Anthropic. Anthropic was trying to sort of like govern what uh their model would do for them, and then ChatGPT or OpenAI sort of stepped in and said, okay, we'll give you a model that does mostly what you want. And uh and fundamentally, uh the government is going to continue to have these sort of uh no restrictions or no holdbacks on whatever models that they use. But then there's this other risk that uh if we get to recursive self-improvement, right, where the models uh become uh so good that they don't need human intervention to make another better model. Uh and then the intelligence level is sort of like not capped. Um, and uh and these models might start, you know, like having goals on their own. Uh maybe the goal might be we need a better, like we need to uh you know, Earth needs to preserve its um environment uh so that it's uh we we don't kill the environment or whatever. But but then based on that goal, it might determine that like maybe it's not good for humans to be around. And then and then it takes action to sort of uh uh end humanity. That is the sort of like fear that some of these AI researchers have, is that it might become so smart that it doesn't necessarily consider humans uh any different than we might consider ants, right? Like we step on ants all the time, it's just inconvenient to take a step around them. So um, if we become ants to these super intelligent AI, then that would potentially be a problem. Now, we're quite a ways away from that because if we have superintelligence, but it's only limited to the data center or it's inside of a computer, we can always unplug it. That's the sort of like unplug argument. But the problem is we're about to put these super intelligent systems into robotics as well, uh, and robotics that could prevent you from unplugging it or robots that could plug it back in, right? So robots that could take action. So if you fast forward this, like let's say 10 years from now, the repercussions of what AI systems might be able to do to humanity as a whole is pretty serious, right? And and um and I want the smartest people in the world to be thinking about these problems. I I am not in the camp that this is all a hoax and it's just meant to slow us down and to uh make America not be able to compete as well. No, this is a real threat. Uh the probability, the 10% probability is ridiculous and stupid, but there is some non-zero possibility of that type of thing occurring, and we can make sure it doesn't occur by having the smartest people that we have working on these type of problems without acknowledging the problem. We won't have anybody working on it, right? So I think it's a little bit nuanced. Uh, I think Dario's um sort of take and sort of raising the alarm bells uh in the way in which he has is uh is just shows his sort of like lack of uh experience being the head of a uh the CEO of this sort of now multi-potential trillion dollar company, suddenly like going out from like never having led a company before to suddenly uh leading one of the most important companies ever in humanity. And he he has like no experience. So he's out there like asking for government to intervene, as if like government knows what to do. Um, no, like government has like one of the sort of dumbest workforces, if you will, in and then you and then oftentimes they're motivated by politics, and uh different administrations change their motives and goals suddenly overnight, uh, change all the leaders of a of a particular department of uh of the government. So, you know, our previous AR czar, AI czar was Kamala Harris, right? So the current one happens to be David Sachs. If you you you know admire David Sachs's viewpoints, just remember that the next one might be a Kamala Harris. So like that that would potentially be the flip-flop. Or if you hate David Sachs and you don't think he should be uh in control of it, then there's good reason for government not to necessarily be in control of it. You want people who are like experts in the field who are um kind of like Elon's way of thinking is more in line with it being correct, which is that you want people who are uh in the field doing the work, checking each other's work, much like you have peer reviews or an editor reviewing the writing of a uh the the the uh a writer's uh novel, for example, where the writer himself doesn't like review and edit his own novel because he's not gonna see his own mistakes. So I do think there is something there, uh, and I do think that the industry is going to figure it out, but you know, we you know, I I'm not in the fear-mongering camp. Um, but I do think that we need to sort of like make sure we don't take the threat or dismiss the threat completely, and we don't we shouldn't take it too seriously as well. So to where we we cripple ourselves.

SPEAKER_04 21:14

Okay, we've spent 20 minutes on our first topic. Let's let's move on to topic number two, um, which is Robin Hood. So there's been a lot of drama here um on Twitter between uh Robin Hood uh Vlad, uh the CEO of AMC, uh Adam Aaron, and then not a lot of people I don't I I think saw this, but even uh Rivian CEO RJ uh had a tweet about this, which he deleted, so I'll get into that too. Um so all the commotion is about Robin Hood's uh stock tokens, where in internationally, instead of buying US stock on Robinhood, you can buy Tokens that are supposed to be backed one-to-one by their stock. Um now, this is a totally new concept that Robinhood is introducing. Um, and with that, there are some potential uh issues, but there are also benefits. So, what Robinhood is really trying to solve is access to the US market internationally, 24-7 trading, and instant settlement. At least those are the big ones to me. I'm sure there's others that Vlad might say. Um the issues that Adam Aaron, the CEO of AMC, brought up is this is being issued without his company's consent. Uh, this can confuse shareholders of what they actually own. And at the moment, if you own uh AMC stock tokens, you have no voting rights for AMC, which Vlad has said that is coming down the line, but today that is true. Then to make this a little bit more confusing, um RJ uh tweeted, and he pretty much agreed with everything that Adam was saying, but he also added on this additional layer, which is you can have uh meme coins or you can make your own token that is in theory also backed by these tokens. So it's a a derivative of a derivative. So there he the reason why RJ, I believe, deleted the the tweet is he called out a specific meme coin that was targeted at Rivian. And that meme coin went up, I think like 400% after he mentioned it, even though it was negative. So like that was not what he meant what if he intended to do. So I I I believe that's why he he took it down. Um but it is uh it is this weird spot to be in because I RJ didn't really say this, but my interpretation was in addition to what Adam said, you don't want your brand to be tied up in this meme culture unnecessarily. Uh and it also hurts shareholders, not just the ones that maybe mistakenly bought the token, but if this is going to fluctuate your stock price, um, it's going to hurt real shareholders too.

SPEAKER_01 24:29

So then it wouldn't fluctuate your stock price any more than buying or selling in a stock would, because under under the covers, essentially Robinhood is buying and selling the same shares that you are buying and selling in terms of the the token. Yes, it's just making it accessible from outside of the US.

SPEAKER_04 24:49

Right. I'm saying if there is some like meme run that is created on on the token and the second derivative, if you will.

SPEAKER_01 24:58

Right.

SPEAKER_04 24:58

And yes, and that's why Robinhood is buying a lot of these shares, and then if there's a sell-off, like it could create these like weird fluctuations, right? Um so then the question is what should Robinhood do? What's the right answer here? And obviously there isn't one, but I'll give I'll give my opinion, which is there's clearly issues, but similar to uh where we are with AI, like I bias towards innovating. Um, and I think the there are real pros here. Yeah, um, it's kind of similar to when Robinhood first came out and introduced no fees on trades, right? No one else was doing it. Well, and I slightly disagree with that, but but keep going, make make make your argument and then thank you, thank you for letting me finish. Uh it's similar to no fees on trades, where Robinhood kind of started that and the rest of the industry adopted it. Robinhood is now pushing for accessibility to US markets, uh instant settlement, and 24-7 trading. I'm hoping the market will this this will be a push for the rest of the market to adopt that. Um now, if I had access to buy a stock token, would I? And the answer is no. Um the the reason being this is totally the wild west. I don't know what I don't know. It's almost as if banks just became a new thing. And like how am I supposed to know what a bank run is without experiencing it? Like, I don't know what the real risks are here. Um, and there's no one checking Robin Hood to see to see if it's actually a one-to-one holding. Um I'm just going through my notes here. So I I guess my point being like there is an element of investors need to be accountable. Um, obviously with the meme stuff, like the derivative of the derivative, that's very obvious, at least to me, that like there's a lot of risk there. The stock tokens themselves, there's in theory is less risk, but we just don't know. Um, so I do think Robinhood should make it as clear as possible what uh these tokens are. Obviously, or maybe not obviously, you know, many people don't read the disclosures anyway, so I don't know how helpful that would necessarily be. There needs to be thoughtful regulation at some point from governments on this stuff. Um, but until then, I think Robinhood should keep pushing forward uh on innovating. So yeah, it's I like the innovation.

SPEAKER_01 27:53

Uh technically innovation. No, no, I somewhat agree with your position. I like the innovation uh aspect of it. I do think that Robinhood is definitely onto something. And and the easy solution here is let the companies sign in, opt in to this, because what company wouldn't want additional shareholders to have access to their shares if they could. It would help increase the price share price of their stock. If all of a sudden their shares are accessible to people in Europe, that's a good thing. So uh company endorsement would actually be a good thing. I think the uh Robinhood could just simply do an opt-in system or allow people companies to opt out. You don't want your shares sold as a token, no problem. We won't sell your shares as a token. Uh the one-to-one has to be verified by a third party so in audits so that they can make sure that each one of the tokens is represented by an actual share. The moment somebody buys or sells the token, that the second that you know, like essentially Robinhood can transact that share uh on the open markets, it would do do so so that like there is no um it has a direct one-to-one uh um sale or purchase, which would affect the price positively or negatively based on that demand. And then if the company wants to do it, there's you you know, like literally no downside in my view. Um but if a company like doesn't want to do it, like a company specifically does not want their shares accessible to some some countries or a particular or any other country outside the US, I feel like the company should have that right, right? Like it they should be able to enforce that rule essentially, rather than some third party uh buying their stock uh with with their uh with the intent to sell it to someone else, but they sort of hold on to it and represent this other uh fourth party, essentially. So um it's it could be a little bit problematic. I I do think that uh Robinhood probably needs to rein it back in a little bit, but still continue to innovate and provide that uh that capability to the individual companies. But you know, it's it's definitely interesting. When you're pushing the envelope, these kind of uh issues come up. So absolutely.

SPEAKER_04 30:12

I do think that's I think that's a really good perspective. The a problem that I have with Robin Hood, and I I love Robin Hood, but a problem that I have is they are just so um, I don't know what the term is, unbiased towards any negative financial consequences for their customers. Like they just want to offer any and all financial products. Right. And and like they they don't care the good or something. Let the customers decide, yeah. Exactly. Let the customers decide. Um, it's similar to prediction markets, and I know there's a lot of other uh places that offer prediction markets, but most of the time you're going to lose money uh in these markets, not make money. Um, so as a customer, you just need to be aware of what you're signing up for when you're signing up for Robinhood, that like you really need to be on top of your own stuff. Um as opposed to signing up for a more standard brokerage like a Fidelity or a Vanguard, where it's only going to be, you know, the most legal. Uh the most legal, the most le the most legal option. I'm not saying any of this is illegal, but uh, but it's gray, right? Like it this is very much on the frontier of finance um where it's still being figured out. Uh okay. Next, Meta Muse. So we talked about this last week, but there's there's been a lot of updates. Uh they I believe are number two in the app store. Their app store rating is a 4.9. Um I did give them uh a little, I gave them a little uh uh negativity last episode because I was saying they really should have focused on the uh professional consumer as opposed to personal. What I totally underestimated is that it is free, and that is going to help drive this immensely. Um what are what are your thoughts on Meta's uh Muse adoption and all the updates they've kind of announced since?

SPEAKER_01 32:21

Okay, so Meta Muse is so good that it actually has me worried about AI more so than anything else, because for the first time so far, I had resisted giving access to any AI, uh access to my email, to any AI. I haven't given you know email access to Claude, to ChatGPT. And I relented and gave access to my emails to Muse of all companies. Um and uh it's really good, it's scary good. It's like I I told it, you know, like I'm waiting for this email from this person. Can you just let me know when it happens? Like, no problem. I'll check your email every 15 minutes. And it's like, wow, it just it just and then like suddenly it alerted me that the email you were looking for is is it just hit your inbox. Here's a summary of it, and it was just like perfect. And it allowed me to respond to something that I probably wouldn't have seen for hours longer. Uh, and I would have had to scour like hundreds of unread messages to find it. Like it and who knows if it would have been there if I was scanning it. So it is really, really good product. And like everyone who's using Muse is saying the same thing that is a it's an absolute like great assistant. It's the assistant everyone has been wanting. Um, and it turns out that Meta of all companies is the one doing this. So the narrative on Meta in the past month has has shifted to from Meta is the tobacco of the tech industry. They they only hurt children, they uh create products that uh are meant to be uh uh addict people and is causing suicide rates to to skyrocket. Um they don't they and Mark Zuckerberg doesn't know what he's doing with respect to AI and uh the metaverse and capex spend and all these AI hires that he spent a billion dollars for or whatever. Uh that was the narrative a month ago, and he's just wasting money. Meta's going nowhere on AI. To suddenly, wow, Meta has one of the best AI models out there. It has the best assistant out there, it uh might be on the frontier. Mark Zuckerberg is on the uh is a thought leader in AI. He's got the most reasonable like perspective on this thing that, like, yeah, obviously security is important, you should be doing so. You don't need other people's approvals before you uh slow down your progress and release products. Uh, make sure that you're releasing products that are safe. Duh. Like it that narrative change is absolutely mind-blowing. And um, and of course, in my view, from a stock perspective, it's still not baked into the stock price today, even though it's already up 10% from where it was uh a week ago. Um, it it has uh already uh added more than $100 billion of market cap. Uh, this this product, the Muse product, has already added more than $100 billion of market cap to Meta's market cap. So um not nearly enough, in my view. And I think Meta now has a real shot to be like the leading AI company. Uh, even beating ChatGPT and Anthropic, I I think it's not going to do it overnight. I think it might take a couple of years, but it has a real shot of doing so in the next couple of years, especially if this level of product quality is what they start to release, and then they get into sort of enterprise and do other things uh that that um uh is on the frontier side. So super exciting. Um yeah. Have you used it anymore since last week?

SPEAKER_04 35:52

Yeah, I've been using Muse and Grokbot. And sometimes I'll give it both the same task to see how it does. Um, I mean, they're very uh comparable. There are some things that Grokbot can do that MetaMuse cannot at the moment. And like it's sometimes it's like an easy one. So, like for example, um, I asked, I was I wanted to to know there was a new AI music model that came out, and I wanted to know what Reddit thought of it specifically. And I asked Muse and it said it didn't have access to Reddit, which is weird because it has its own computer, so it should be able to just go on there and and read it. And Grokbot did it, no problem. So there's still there's definitely some small things that they need to work on. Grockbot's not perfect either. I've run into issues with them as well. Um but the the one thing, or I guess the two things that kind of give me a little pause in the sense of like Meta is not out of the woods yet, is one there is clearly a a lot of success with this new type of agent that is in the cloud, has access to its own computer, you could set up routines really easily. The UX is a chat interface, like a messaging interface as opposed to just like chat. Um we're gonna see Anthropic and OpenAI do this at some point. So that'll add a lot of competition. And then the other piece that gives me a bit of pause is this feels very similar to the initial success that Threads had from Meta, which is their Twitter competitor.

SPEAKER_01 37:31

There's a big difference, but but but finish that.

SPEAKER_04 37:35

No, go. What's the big difference?

SPEAKER_01 37:39

Okay, so with uh with threads, there was already a leading product, X, uh, you know, Twitter, that uh had already captured the the uh all the people. If you think of the sort of like S-curve of the market that we talked about last uh last week, uh basically the S curve was on the top of the S curve where like everybody who wanted a Twitter-like product already had one. It's called Twitter. You just use Twitter, right? Like so um they they weren't, they were there were late entrants into a market that already had a leading player. Uh ironically, usually markets sort of like have at least a couple of players. There isn't two players in that market. Uh, there's basically Twitter, and there is this sort of like additional benefit that comes from having a single town square that everybody talks in in as opposed to multiple town squares. So the way in which Thread has competed is by essentially getting their Instagram user base and Facebook user base to uh pushed to get onto Twitter or to Threads rather. Um, but fundamentally, those people are getting pushed to get into threads, and they're not natural users of threads or that type of product. Uh, and therefore, if despite the large user base, the engagement is really, really shallow. Uh now, when we contrast that with uh an assistant, an AI assistant, as opposed to uh a chat chatbot, which which chat GPT essentially has captured the consumer on the chatbot side, meaning you have a question, you ask ChatGPT or you ask Google in the in the um in the URL prompt. But uh in to have an actual assistant that is doing work for you, reviewing your emails, checking your calendar, that sort of thing, people are not yet using ChatGPT for or Cloud for. Um, there are some users, no, don't you know, don't get me wrong, but the masses haven't come on board yet. And that's the big difference here with uh with Muse. And then the product is getting rave reviews and is doing it really well. Uh, and then they have huge distribution capability of being able to push this product to 3.6 billion daily active users, hundreds of millions in the US alone, uh, which is sort of like the market that they have started in, uh, which is gonna allow them to um basically become very quickly the leading player, and then everybody else has to play catch up. They already, despite um Grokbot coming significantly before them, I think Grokbot came out what, like at least a month before uh Muse did, right? Grokbot bot is essentially one of the new entrants into this space, but um uh and it immediately captured some market share of the assistant world. Uh and then Muse already has what seems to be maybe two or three times the distribution of Grokbot, uh, just based on like the number of reviews on the app store, they have more than that many times uh the reviews, and they're both getting rave reviews. So uh on a you know quality basis, you can sort of like consider them on par with one another. But in terms of distribution, uh it turns out that Muse has way more significant distribution capability with Meta.

SPEAKER_04 40:49

But I I believe that's more so because of the price of it being free. If GrockBot was free, there would be way more downloads. I mean, Meta has a bigger distribution channel, but X is still a huge distribution channel, right?

SPEAKER_01 41:02

Um that's very true. That's very true. Yeah, and I think that's what Yeah, I agree.

SPEAKER_04 41:07

We're in terms of the S curve, we are very much in the early stages as opposed to the later stages. But if ChatGPT comes out with a product like this, I mean they're the number one player right now. So I just I don't and and users are quickly changing products. So like yes, I'm just concerned that Meta's not going to be able to hold on to this uh a little lead that they that they seem to have form. Um, I'd love to be wrong about that, but it's it's just a tough space, right? As a shareholder, I'd love to be wrong about it. But um those those are my concerns.

SPEAKER_01 41:45

So yeah. I mean, that's a definitely a good point. This is not a uh done deal by any means. Any it's still anybody's game, it's still so early in the AI world because the masses haven't yet are not yet using you know some of these agent and assistant capabilities that um that anybody could still win this game. Yeah, and I believe Google, by the way.

SPEAKER_04 42:07

Oh, I mean, I still believe Google's going to be like the most prominent AI product out there. But um, on the flip side for for meta in terms of positives, like as a shareholder, this is exactly what I want to see. That they're making killer products that uh are connecting with users that are extremely capable, that are providing new value. So they're checking a ton of boxes and validating the thesis as to why I think both of us uh are excited about Meta. So this is all good. I'm just not I'm not you know waving the the flag yet that this is over.

SPEAKER_01 42:46

Um, by the way, they did something else that uh was super interesting, and I didn't get a chance to look into it, but they um announced Meta One uh subscriptions at and I think a basic subscription is like $8 a month, and a uh premium subscription is $20 a month. And and those subscriptions give you not only more usage of Meta Muse, but they give you Instagram Plus and Facebook Plus and WhatsApp Plus. And I don't know, I didn't look into like what does that plus give you, but there's some additional features that they give you in each of their existing products that people might want to subscribe to anyway. So that they're they're now in this position. Uh, I mean, like this is this is a very good comparison to make, which is that if you remember, Zoom was like by far the number one video you know conferencing app uh before COVID. And it you know shot up during COVID and it just like this was this rapidly growing company. Um, and then Microsoft decided with an inferior product, Teams, to just include it as part of the subscription of their Office 365. And all of a sudden, hundreds of millions of uh Teams users had access to Microsoft Teams, and Teams has overtaken. In Zoom, as this sort of like more usage of a product that is very inferior to other products. So the distribution of meta gives it a huge advantage. So distribution is like very, very important. And then product quality is obviously very, very important. You still have to have a really good product, which in Muse's case they do so far. But even if, for example, ChatGPT or Anthropic were to come out with slightly better versions of these agentic systems similar to Muse, but they don't have the same distribution, that might be a challenge for them. Now, ChatGPT has pretty good distribution. But still, Meta has 3.6 billion users. So it's still a three three and a half time larger distribution uh platform.

SPEAKER_04 44:49

Yeah. One thing, uh one more thing I'll say going back to Google, absolutely killing it. Part of the fear, and you were kind of uh alluding to this before, where like you don't want to give AI so much access to your stuff, most of our stuff is in our email. Who already has access to our email? Google. So if Google created a product like this that just had access to your Gmail already, plus all the other Google services, like the there's no barrier to entry as opposed to Grothot and Meta. There, there's this just second thought of like, do I want to give it access to this? Google doesn't have that. So yeah, just another angle of it.

SPEAKER_01 45:36

Yeah. Google is in a great position to do it. It's just that they're I feel like they're asleep at the wheel. They need that founder, uh, founder to get back in there.

SPEAKER_04 45:44

Yeah, but uh, I mean, it it I think they're just due for a model refresh, right? Like when Gemini 3.5, I think it was 3.5, was like initially released, or maybe it was 3.0, like it was a very strong model, and then it's just kind of aged and they haven't, you know, it's it's what happened. So I'm confident there will be a newer, better model, but I agree. I wish they had stronger leadership. Um, and that is definitely another thing that Meta has going for it. Uh okay, let's talk about Bumble. I know you sold your Bumble, but I didn't, so we're gonna still talk about it.

SPEAKER_01 46:23

Um so it came out reminder every every time as the stock goes up.

SPEAKER_04 46:31

Um so it came out, I think this was last week, maybe this week, uh, that Blackstone is going to fully exit their bumble position. Blackstone has been an investor in Bubble, Bumble for the last seven years. Um, they've been selling since around 2025. So this isn't like something new necessarily, but it is new that they are going to fully um exit. So the downside here is you could infer that Blackstone is not a believer in either Bumble's turnaround or acquisition. Um if there is going to be an acquisition, if you could recall, there were rumors that Bumble might be acquired um a couple months ago at this point. Uh, so that isn't great. But I don't normally uh invest based on what Blackstone is doing. So that could be disregarded. And the positive side as an investor is now you have this selling pressure that is going to go away. Um, the estimate is they're going to be done uh selling by early 2027 because they have to sell in increments. Um, so that's the silver lining here. But I I'll be honest, it does give me a little bit of pause that they are exiting because this is a sign that they that they don't believe in what the company is doing. So yeah. What what are what are your thoughts?

SPEAKER_01 48:00

I mean, you I mean it doesn't change anything for me. I've known that Blackstone is uh exiting the position since I started becoming a shareholder. And um I I was in it because I thought that Whitney uh was going to turn the company around and rapidly uh improve the product and release a new product, and none of that has happened. And I still go into the product every so often just to check it out to see if anything has changed, and it's like, oh my gosh, it's still painful to see it be basically the same product that it was two years ago. Um, so I'm still disappointed, and of course, you know, uh I sold out all my shares, so I like most mostly it's out of mind, out of sight, um, right, out of sight, out of mind. But um, but yeah, I'm I haven't seen anything to change my mind about and get more sort of like uh bullish on Bumble yet.

SPEAKER_04 48:51

So yeah, I for me, and I've said this time and time again, like it is strictly about this product overhaul. I want to hear what the details are, so then I could judge for myself. With that said, there is, and I think I've said this before on the pod, there is a world where I do mitigate risk and sell some shares, um, probably for tax loss harvesting uh before the end of the year, but I do expect to hold on uh unless if there's a really negative update about this overhaul. So okay. Uh with that, let's get to listener questions. Adrian, what do we have?

SPEAKER_00 49:28

We've got a lot of questions, a lot of good questions.

SPEAKER_04 49:32

Good, let's do it.

SPEAKER_00 49:33

Um, okay, let's go ahead and our first. Well, let's since we were just still talking about Meta. Um what is someone is asking? Let me pull up. I think this is a good question for Hamid. Hamid, I'd like to hear your own take on the strongest bear case for Meta. Thanks.

SPEAKER_01 49:59

I mean, the strongest bear case for Meta is that their existing products are gonna decline in usage, right? Because um because of the additional restrictions that they're putting in for uh one and two hour limits for teenagers, and therefore the usage of the products might go down. Uh, they're also reducing the notifications during school hours and at nighttime for uh for teenagers, all the way up to 18 years old, by the way. So there's that. Uh there's the uh, you know, like uh the the AI product uh spend and uh uh talent spend and uh CapEx spend might not materialize. I mean, Muse has a good early start, but you know, like Dustin said, uh ChatGPT could potentially crush it, uh Anthropic could potentially crush it, and if they don't continue to have a great product in the AI world, so that might not materialize. So the combination of them not having any new products in the AI world that actually generate revenue combined with a decline in usage of their pre-existing products, which would imply decline in advertising revenue, uh, would be the bear case for it. Um, I don't think either one of those things are going to materialize. So I'm not a bear on it.

SPEAKER_00 51:19

Dustin, do you have a bear case?

SPEAKER_04 51:22

I feel like I laid out not necessarily a bear case, but just my concerns um of the pitfalls ahead in this AI race against other uh companies. I mean, even if Meta didn't wasn't in AI, they still have an uh extremely strong business. So um, yeah, I don't I don't think there's necessarily that strong of a bear case. They're in a lot of different products and services.

SPEAKER_01 51:52

So by the way, we we didn't talk about this. I kind of wish we had during the meta uh segment, but uh MetaConnect, uh their conference where they announce new products is about to come out. And they've they're like announcing products left and right in the past month before the MetaConnect uh event. So it'll be super interesting to see what they're saving up for the MetaConnect event. I'm actually quite uh quite excited for that.

SPEAKER_02 52:17

That's happening next week, you said? Uh yes. Um okay, we'll talk about them next week.

SPEAKER_00 52:24

Okay, one more about Meta. Hi, Hamid. Thoughts on Meta One, the new paid subscription service for Instagram and the upcoming Meta AI glasses event. Do you think these are already priced into the stock?

SPEAKER_01 52:38

I feel like nothing is priced into this stock because just based on their existing businesses, this stock should be, uh, in my opinion, an eight or nine hundred dollar stock. And if you take the potential of what they're doing in AI and uh VR AR smart glasses uh into consideration, this should be over a thousand dollar stock. That would be my opinion, obviously, not investment advice. Um, but yeah, I I think all of those things are upside for it.

SPEAKER_00 53:05

Okay, this is a good one. What happens to Google's advertising revenue if the only people Googling stuff will be AI agents who don't care for ads?

SPEAKER_01 53:19

Well, it it's a pretty major assumption to assume AI agents wouldn't care for ads. Uh, it's quite possible that they would care for ads. Um and uh advertisements is how we discover things. So why wouldn't AI agents also discover things through advertisements? So there's there's that factor. Uh and of course, these types of um transitions are never quick, meaning, like, it's not gonna go from people Googling stuff to 100% to AI agents Googling stuff 100%, right? It's gonna even if that was going to be the bulk of um searches in the future were gonna be AI agents, it would take dozens of years before it goes from 100% from one to the 100% of the other. Um, so yeah, I mean it's it's gonna transition over time. And that assumption that um AI agents wouldn't care for ads, it might be a wrong assumption.

SPEAKER_04 54:16

The the other, I agree with that. The other thought is revenue, the where revenue is coming from is gonna change, right? Like it yes, Google has free AI services, but there are a lot of Gemini features that are behind a paywall. Um, and similar to Meta with their Meta One subscription where it's buckets a lot of things together, Google's doing the same thing where you can get advanced Gemini features, you can get uh more storage for your Gmail and Google photos, you can get uh YouTube premium all in one package and more. Um, so I think over time we're going to see more revenue come directly from uh their users, which I don't think is necessarily a bad thing. And they'll still have the advertising business.

SPEAKER_00 55:05

Okay, let's move on to more talk about Rivian. What do you think about cities' neutral rating for Rivian? And would you purchase them at the current levels after today's sellout?

SPEAKER_01 55:16

Well, today's sellout probably happened because of the uh interest rates going up a quarter percent. Um interest rates going up generally hurt car companies because a lot of their customers are um using loans. And the idea here is that uh fewer people would buy their cars because uh it's more expensive as interest rates go up. Uh now in Rivian's case, because of the bulk of what they're building is um uh the R2s right now, and the R2s are uh are maxed out from a supply standpoint as opposed to uh the people who want to buy one. Um, meaning you can't just go buy an R2 right now and and just pick one up, uh, they're not going to be affected by interest rates in any significant way. So I like I consider these sell-offs uh at because of macro issues as like non-issues from a long-term perspective. I still continue to be bullish on Rivian, but fundamentally, Rivian's fate might not change until it starts to show rapid revenue increase combined with rapid uh you know, positive margin increase as well, which neither of which they have shown yet. But that's all, in my opinion, about to change in 2020, the second half of 2026.

SPEAKER_04 56:37

Rubian's only down like two percent today, so I would wouldn't even consider like nothing could have happened in Rubian could have been down two percent.

SPEAKER_01 56:44

And down two percent.

SPEAKER_04 56:45

I thought it was down more than that at one point, but maybe at one point, but it it ended the day only being down two percent. The market, uh, from my perspective didn't really move that much uh with the rate hike, which was the most probable outcome. And I think majority of the market was expecting it. A lot of people on Twitter, I don't think, were um, but uh yeah, I think the market digested it pretty well. At least today, there's always a chance.

SPEAKER_01 57:10

It bounced back up towards the towards the close. Interesting. I I just noticed that a lot of stocks bounced back up towards the close after going down heavily when the interest rate cuts or increases happen. Yeah, that makes sense. Okay, what's next?

SPEAKER_00 57:26

Okay, we ready to talk a little bit about micron?

SPEAKER_01 57:29

Let's do it.

SPEAKER_00 57:29

I know we love it. Um, there's a few things that we we'll talk about, but here's the first one. The labs are gonna have to dedicate a lot more compute to safeties. Does that make my or does that make you more bullish for micron?

SPEAKER_01 57:45

That seems like a weird take. Uh the okay, so the the let's talk about why people thinking um if the if the labs are gonna slow down AI, that's a negative thing for uh semiconductors and micron in particular. The reason is because if you're not training new models, then you need uh less GPUs, and those GPUs would be available for someone else to be training models that and and therefore uh it makes GPUs as a whole more available. And if GPUs are more available, then you don't need more of them. Um and uh if you don't need more of them, then you need less memory, right? Like because each GPU needs a bunch of memory to go along with it. So less training of new models would fundamentally mean less silicon for everybody, you know, whether it's Nvidia or Micron, etc. Um, but uh there's another set of uh growth that is also happening, which is the usage of existing models. The usage of existing models is growing exponentially, and that also needs GPUs and it also needs memory. And therefore, we can't even keep up with the usage part. So even if we were to just stop uh new model development completely today, I don't think we're gonna have excess GPUs or memory to begin with. So no matter how you look at it, I feel like we're very bullish on, or I personally am very bullish on uh Nvidia and Micron. Uh obviously I'm more bullish on Micron because it has a lower uh price from a per dollar of profits perspective. But um but yeah, I don't see the slowdown in AI actually slowing down silicon anytime soon. Uh and that's assuming that any slowdown in AI happens. You know, these guys are still going to be moving forward. They're already talking about each each uh lab, by the way, is already talking about their next generation of uh models that are coming out. So the slowdown from some people's perspective might be very different than the slowdown that these guys are are are talking about.

SPEAKER_04 59:56

So yeah, I so I I agree that um I don't think the increased safety is going to increase uh memory for these frontier labs, but I'll play the devil's advocate of let's say the new AI models are so powerful that they need like a constant AI, like I don't know, uh going against what they're trying, you know. Like there just needs to be something constantly running to block the AI models. That that would require some memory. I don't think that's going to happen. But that that's the only way.

SPEAKER_00 1:00:40

Okay, continuing with that. Hamid, what do you think of the choppy and downward price action today for Micron and Sandisk compared to the rest of the AI semis that outperformed?

SPEAKER_01 1:00:51

I I don't even bother like thinking about it because I don't I can't predict the market on a day-to-day basis. It's it's just irrelevant, in my opinion.

SPEAKER_00 1:01:01

You have anything, Dustin?

SPEAKER_04 1:01:03

Micron ended the day down 0.11%. So I again I understand it was choppy during the day, but it was pretty flat where where it landed.

SPEAKER_01 1:01:15

And you know, like Meta, in my view, Micron deserves to have a price point that's more than $1,500, maybe $1,800 a share, maybe closer to $2,000 a share. Um meta deserves, in my view, a uh stock price that's over a thousand dollars a share uh based on where they are. So uh the market doesn't care about my view, just in case you didn't know that. But but uh I also don't care about the market's view. I kind of care about my view.

SPEAKER_04 1:01:44

So that that's what that's what creates opportunity, right? Like if the market only followed your view, you wouldn't you wouldn't be as successful as you are in investing because there wouldn't be that opportunity of finding where the market is because I'd be like, oh, it's exactly priced like how I think it should be priced, and then therefore why would I buy it?

SPEAKER_01 1:02:02

Um but but yeah, the the stuff that I buy is because I I feel like it's undervalued relative to how the market makes its value.

SPEAKER_00 1:02:12

All right, micron still. What do you think would change investor mindset of Micron's forward PE to be more similar to typical forward PE's of the Mag 7?

SPEAKER_01 1:02:21

Consistency, right? So if they can consistently deliver uh revenue and profits uh at some base level that uh hopefully um gets set at in 2027. Uh, you know, if 2028, for example, grows that revenue and and profits from the 2027 base, I think the the current model expects them to go down. That's the reason why they're trading at a forward PE of seven, is because uh the market is assuming that it's unsustainable. But if they can continue to sustain uh those uh revenue and profits, then the market has no choice but to say, you know, why would a dollar of profits from Micron be viewed differently than a dollar of profits from Nvidia? It just doesn't make sense. So, you know, at some point, if it if the consistency is there, um they'll they'll they'll get their uh PE ratio. And by the way, they could continue to get a PE ratio of seven and still increase pretty rapidly just because the E part is growing, right? Like uh uh you know in 2027 they're gonna have significant, greater, significantly greater uh profits. 2028 is expected to have significantly greater profits as well. So, you know.

SPEAKER_04 1:03:40

I think um the consistency is the number one thing. What will also help them is share buybacks, and I would love for them not necessarily to do acquisitions, even though that would be fine, but invest in other companies um as well, because they have so much cash on hand, uh kind of similar to what Nvidia is doing.

SPEAKER_01 1:04:01

Yeah, yeah. So so for Micron in particular, they haven't had the cash yet, but they are about to have like a huge influx of cash um from this quarter on.

SPEAKER_00 1:04:13

So okay, this is a good one. Following up on the S curve from last week, where would you locate Micron, Rivian, and Rocket Lab?

SPEAKER_01 1:04:27

Well, not everyone who wants memory right now can get them their hands on memory. So that that's uh that that's the bottleneck for uh for for micron. Um so the the memory market uh prior to AI was already on the right side of the S curve. AI is what has changed everything. In fact, the GPU market prior to AI was already also on the right side of the S curve. But but you could almost think of the uh AI as creating a new S-curve that just started from scratch for GPU and memory demand. And what I would say is we're somewhere on the exponential curve of the demand. And when it'll slow down and get to the top portion of the S curve is sort of anyone's guess. It all depends on how useful AI continues to be. And if it continues to be useful and it actually grows the GDP and uh the companies that are investing in AI just continue to invest, we might see that exponential growth for another decade. It's kind of hard to predict exactly. Um people, you know, have been estimating that uh it was uh for for three years they've been saying that uh we're putting too much money into AI and it's gonna, it's about to like uh the bubble is about to burst and we're we're gonna reach the top. So um again, it's kind of hard to predict. Now, on Rivian side, the market is very well known. Uh the the car market. The car market is already on the right hand side. So we know what the market size is. Everyone who wants a car has a car today, right? But we also know a few other things. We know that 100 million people change their cars worldwide every year. So uh there's like um uh you know something like a billion cars worldwide, maybe maybe two billion. I don't know the exact number, but about a hundred million of them get changed swapped out every year. Uh and that number is dropping, by the way, because cars last longer and EVs last even longer. So maybe that number becomes 50 million over time. Um, but in my view, the disruptive technology that um that EVs have created as well as autonomous vehicles have created is going to make all cars in the future be EVs and autonomous. So that curve is just on the early side, and we know exactly how big it is because we know how big the car market is. Um, and therefore we're like literally at somewhere around 8% of the market penetration on that S-curve. So we're on this exponential growth curve, but only at 8%, and we're gonna go all the way up to 90% before it slows down. So that's where I think Rivian is, or Rivian's market is uh in terms of the S-curve. Uh and then Rocket Lab is another interesting one because not everyone who has access to space has access. In fact, it has been out of reach for everyone who wants to put satellites into space because of how expensive it is. It's it used to cost thousands of dollars per kilogram to put a satellite into space. That cost has been uh coming down very rapidly. Um and uh and then humans want to go to space, they want to do space tourism. So if you could get the cost down, this exponential curve is going to grow for many decades. And we're just like at the very, very early, like on the on the uh S-curve of space, I would say we're on the very left side where companies like SpaceX and even Rocket Lab were on the very early, really hard uh part of the S-curve where they're not yet profitable from their space businesses, essentially, their launch businesses. So um we're gonna get to this rapid exponential growth curve, probably somewhere around five, maybe 10 years from now. Um to where, you know, when you can take a uh 10-day vacation on the moon, that's when we're on the sort of like right side of that S curve. That is probably decades away. Um, so Rocket Lab and SpaceX, from a growth perspective, they have a lot of potential. It's just the the matter, the question is how quickly can they get there? Uh and I think Elon Musk is way too optimistic in his numbers of trillion dollars of revenue by 2030, you know. But fast forward another five years after that, and maybe okay.

SPEAKER_00 1:08:56

Let's do this one, and then I think we can close from from here. Hello, Hamid and Dustin. How do you identify any stock and what are the usual sources of information? How do you decide that you will invest in these stocks or not?

SPEAKER_04 1:09:11

Dustin, I'll let you take it first. Yeah, I mean, so I I've talked about this before. I don't spend too much time looking for new opportunities because I'm so busy working on fantastic products at Savvy Trader. I just don't have the time to find new stocks. Plus, I have two little kids at home. Um, so a lot of it is in what a meet brings to the table, which is why our portfolios are so similar. Um, and then I do like to use our own products to identify opportunities. That's how I found um Iron, for example. Um what I judge companies on, um I definitely look at revenue growth um more than anything uh in terms of the fundamentals. I also, Hamid, cover your ears. I do like to look at the price history to get a sense of if I am buying at a good or a bad time relative to where the stock has been. Um I obviously care a lot about what the company actually does and if I buy into the the product that they're building or the service they're providing. Um and I also that I this isn't a requirement, but I do love stocks that have been beaten up or that are uh are not recognized for the the value that they're providing. So for example, I bought uh Robin Hood when it was really down after all the GameStop craziness. I bought Meta when Pat was really down when everyone was hating on Mark Zuckerberg more than they were now. Um, like I think those are the best times to buy is when like you could kind of see past the market or the fear, which I think is very similar to where we are with um AI, by the way. In the not in the sense that like all these stocks are up, but there is a lot of fear in the market that I think you can kind of see past. Okay, that's it. Your turn, Hamid.

SPEAKER_01 1:11:17

I think all of those are uh excellent. I mean, for me, it's uh some some of those things that Dustin just touched on. And I basically stay with focused within my realm of expertise for the most part, especially for larger bets. Um, and I happen to have grown up in tech, so I know tech better than anything else. Um, so that's sort of like my realm of expertise. So when a great opportunity presents itself in healthcare or um medicine or something that I don't know anything about, I don't have FOMO in just like passing on those opportunities and I just stick to what I know. But yeah, um that's about it. Thank you very much, everybody. Dustin, do you want to wrap uh do you want to close it out or just saying bye?

SPEAKER_04 1:12:02

I'll close it out. Thank you all for listening for the last hour and 12 minutes to us rambling about nonsense. We appreciate it. I hope you liked our new bar at the bottom, which I'll hide the question so you could see it again. I know everyone uh liked it. If you have any suggestions on how we can improve the pod, let us know in the comments below. We appreciate you all. Awesome. Bye.