Above Board

Paul (00:01.346)
Hello, hello, hello. This is the Aboveboard Podcast from Fathom Analytics. am as always Paul Jarvis and I'm joined as always by my co-founder Jack Ellis. Good morning, sir. It's earlier than usual.

Jack (00:14.008)
Good nerd.

Jack (00:17.644)
Yeah, it is actually. forget, yeah, two hours behind. It's 7am for you. That's rough.

Paul (00:20.78)
Yep. Fucking Canadian Thanksgiving, Like, I need to go return something to an actual store. I went there yesterday because it said their holiday hours were 12 till 5. Get there at like 5 after 12. Store's closed. A bunch of other people like wanting to go to the store too. And so we got to go back today all the way downtown, which downtown for me is like an hour away.

Jack (00:37.549)
Mm-hmm.

Jack (00:43.438)
think downtown for every- but downtown for everyone is always a grind, unless you live downtown. Going downtown.

Paul (00:48.43)
Yeah, I lived downtown. I live downtown Vancouver, like literally in the very center of downtown Vancouver for 10 years or so. I remember driving like two or three times a year because I just like, why would you drive in Vancouver if you don't have to? Yeah.

Jack (00:58.019)
Yep.

Jack (01:06.08)
You are downtown already. I wonder if that song downtown is royalty free, know the downtown. No, okay. Yes, this is true. WordPress will find a way to sue us.

Paul (01:12.012)
Yeah, I doubt it is. No, don't sing it. Copyright.

Exactly. There's some breach of open source singing.

Jack (01:23.576)
Use the word word.

Paul (01:25.294)
We don't need to talk about it, but obviously everything's just salty with that. It's funny. I guess I want to talk about why Google search is no longer an actual search engine.

Jack (01:28.686)
Cheers.

Jack (01:33.856)
Yeah, absolutely. All right, so where are we going today?

Jack (01:45.132)
Yeah, it's a mess.

Paul (01:48.702)
So it's funny too, because my wife is like the least technical person in the entire world. And like, she just doesn't understand. Like she tries to search in Google. She tried, so she tried duck duck, go duck duck, go as garbage as well. But then she's like, I'm just going to switch back to Google. Cause I remember Google being good like years ago, she tries to find stuff in Google. Can't find anything. So I'm like, maybe you're doing it wrong. So I tried to find some stuff in Google. Cause I don't use Google. I think I use brave now. That seems to be the

Jack (02:17.71)
Okay, and DuckDuckGo, my understanding is they use Bing and they put adverse. How does it work? Doesn't DuckDuckGo just use Bing?

Paul (02:18.562)
Best.

Paul (02:26.624)
Maybe and Bing is kind of garbage for similar reasons too. So I just find that the results have been getting worse and worse for everybody that I talked to about this has kind of the same feeling that it's just like, I don't know how to find stuff in fucking search anymore.

Jack (02:44.898)
Yeah, you can't. So, okay, so that's interesting. And it depends. So if you're searching for technical stuff, which a lot of people don't do anymore, they'll go to AI. I have a love hate relationship with AI. think perplexity AI does good, but you can't ask chat GPT for technical questions when you're doing anything above the basics of a certain topic. You know, like it may the chat GPT becoming the search engine for the work that I'm doing. And I've made a big thing on this and like, there's kind of a bit of gatekeeper ink.

Paul (03:09.091)
Yeah.

Jack (03:14.942)
gatekeeping language because I'm my position honestly, if you're doing anything complex, AI can't help you. And that sounds kind of arrogant or not. don't know. Like I say gatekeeper is kind of the word there because I'm like, you must not be doing anything complex. And it's tricky because I sort of am saying that I don't like and that's another thing. The reason people say to me, but AI was not no one saying AI can do that. it's the problem is that people are.

People are saying it's going to be replacing developers. already is to placing developers. The only developers it's replacing is people that are writing basic stuff. You know, I put in my work. it's not even close. You know, I even, loaded our entire analytics, let's call it analytics engine. So I gave it full context. This was in Claude AI. Cause everyone's telling me, Claude is better than chat GPT. I load the full context, like all of our analytics files, schema, everything. I don't, it's not.

Paul (03:42.039)
Yeah.

Paul (03:50.944)
So I can't write an entire app in AI yet. That's what you're saying.

Jack (04:12.258)
sensitive information is like, whatever. I load it all in and I'm like, they've got all the context. They can see the class references. They can see this. And it builds me something which then completely breaks everything. And what it had done is it tried to clone a instance of a class, but it completely missed the fact that it was also injecting objects into this class. And when you do a clone in PHP, it doesn't do deep cloning. It's a shallow clone.

Paul (04:13.292)
Yeah.

Jack (04:39.052)
So then you end up with a cloned object with a reference to the old objects. And this is just basic stuff that you would not miss if you were doing this yourself. maybe you would like the more, to me, cloning an object, I know that if I've got objects in the class or in the object, I know that there's going to be a reference done to it. And AI completely misses that. I'm going, what do mean AI is replacing developers? Maybe for the people that are just writing crud. Sorry, when I say crud, I mean...

Paul (04:47.117)
Yeah.

Jack (05:06.446)
create, read, update, and delete apps, basic stuff. So I kind of have a do. I am a bit smug, I suppose, on it. I'm just not convinced.

Paul (05:09.191)
Hmm.

Paul (05:16.75)
Yeah, I mean even for the stuff that I do with AI for code, which is a lot less. It's just like. Generating ways to create events and fathom like if I'm if I'm doing support, it does it wrong half the time and then I tell it like hey, this didn't work or this didn't. This this gave me an error in my console and they're like, actually yeah it did. Here's how. If you knew it was going to do that, why did he give me the wrong and I have friends who've done who've asked it to do like.

Jack (05:27.704)
Sure, yeah.

Jack (05:36.907)
I hate that so much.

Jack (05:42.584)
Okay.

Yep.

Paul (05:46.102)
And it's done math wrong, like simple math, not like multivariable calculus math, like just simple math.

Jack (05:52.43)
Yeah, the other day I was looking into some infrastructure level stuff, right? And the start point was a feature of CloudFront. And the actual piece was, I was trying to figure something out. And the thing I was trying to solve was a certain, I forget the specifics, effectively, wanted to know when I already knew, it's funny, I didn't want to know this. threw me on this pathway because it suggested it. I was doing something along the lines of CDN and at what point does

can I do something? And it comes back to me and it says, CloudFront is actually, CloudFront Edgeworkers, which is what we use for our marketing site, Edgeworkers, CloudFront functions, sorry. Yeah, that's hit before the web application firewall. So you can totally run a function before it hits your firewall. And I'm going, I'm like, what? How does that make any sense? The firewall is supposed to protect it. But they're like, no, no, no. And they're like, yeah, Amazon documented this.

So I like go down and I do a bunch of testing. going, this is amazing. I can't believe this is possible. If you just do this, then it's told me to do. I do all these things. It's not working. I'm going, it's not working. It keeps telling me to try this, try this, try this. Okay. So this is where I'm at. And of course it's just me that's prompting it wrong. Not the fact that AI is leading me down this rabbit hole that, know, intellectually I know is the wrong rabbit hole, but I want to believe it. And I get to the point and I'm going, it's not working. go, are you absolutely sure that CloudFront functions comes before WAF? And I go,

you know, there always seems to be a misunderstanding. go, fuck you. It's not even a computer I'm getting mad at. But just, and then someone goes, you're prompting it wrong. No, I didn't prompt it wrong. It gives me this completely falsified solution and then sends me down this wishful thinking rabbit hole. So no, dude, let's get back to the search engine topic. I'm done with AI for code. It's not for me. I will use it for some basic stuff. And when my expectations are set,

Paul (07:22.604)
Yeah

Paul (07:41.057)
Yep.

Jack (07:47.63)
Okay, this can be basic things. can transform texts. can, new language. I don't understand. It can help me modify code, add code. But for most of the stuff I'm doing, it's not going to help me in any way. And it's going to lead to waste of time. So I'm back using search engines and I'll hit stack overflow because then at least I can understand the concepts and I can upgrade my brain. I don't want to atrophy my brain. AI can help you do this. Okay. Well, you're just going to get worse and worse at writing code. My brain, everyone's brain has this.

Paul (08:07.085)
Yeah.

Jack (08:15.776)
Effectively, I'm not going to say infinite capacity because it's wrong, but it has huge capacity for knowledge and information and connecting dots and everything else. I'd much rather read a concept on Stack Overflow, understand their problem, understand the solution, and then boom, that's now in my brain. It's not something I've just relied on AI for.

Paul (08:33.826)
Yeah. And I mean, so that brings us back to why Google search engine sucks. One of the reasons is that it relies on AI. They incorporated LLMs into results. So this is actually bad for several reasons. So a lot of times, like we just talked about, AI text is just straight up wrong, but it's confident in its wrongness.

Jack (08:44.27)
Jack (08:55.447)
Yeah, it is.

Paul (08:56.776)
and it's Google's AI. remember how Google AI had like, asked it to generate an image of like Nazis and it showed you like black Nazis. Or you asked it to generate an image of like the founding fathers of America and there's a bunch of women. This is like historical facts that are just completely incorrect. Yeah. And then, so that's the first problem is that AI is just wrong all the time. Fight me on this. I don't care. But the second problem is that the point of a search engine or the point that

Jack (09:09.603)
Alright. I remember this. This was so funny.

Paul (09:25.55)
Google said it search engine was and that all search engines are supposed to be is that you find information on the internet. Right. So if Google is going to replace search results with AI, then it's not going to show it's not going to show other sites. It's not going to show sites where that person that human has like shown the work. Right. Or like shown why the answer is the answer. It's just a little blurb at the top. Like, you were looking for this. This is what we think that means.

And so it seems like Google is trying to replace showing external sites. Other than ads, like they want to show ads and they want to show, like AI generated tax, which is just, yeah.

Jack (10:04.814)
they're competing with perplexity. They know that people want and this is this is actually a theme by the way, people outsource their thinking. And I've obviously been in, I don't want to say I've been in the health space, but I've been around a lot of health stuff. And I'm seeing it people just completely delegate their thinking to others. They put trust in those people, but they don't actually want to understand themselves. Granted having time to look into it and everything else is a bit of privilege, right? I mean, absolutely is. But

Paul (10:20.113)
Hmm.

Paul (10:32.248)
Yeah.

Jack (10:34.38)
they don't want to actually understand it. And I think that that shift, you can argue, it makes me move faster. Maybe you're moving faster with the basic things, but, two seconds, two seconds, two seconds.

Paul (10:45.154)
I wouldn't want to move faster in the wrong direction.

Paul (10:52.916)
Ahem.

Jack (11:05.39)
don't know if it was picking it up. Terry was outside vacuuming over my door. So I was delegating a thinking.

Paul (11:14.166)
moving fast in the wrong direction and I said I wouldn't want to move fast in the wrong direction.

Jack (11:19.852)
Yeah, and I think that this move away from just actually looking to understand towards, I just want to get it done. I don't think that that's right.

Paul (11:28.664)
Well trust me, I'm the science dude.

Jack (11:30.87)
Yeah, well, that's another example. And this is a beauty too. So actually, in my circle now, I'm speaking to multiple different scientists and different backgrounds, really advanced backgrounds. And just hearing their perspectives on different things is fascinating and hearing why they believe it. And then what I'm finding talking to people is that each scientist will have kind of gaps in their knowledge to the point where I know more about a certain topic relevant to the science from another scientist or from another clinician.

that they don't even know the basics about. this, like this, this exists in business, by the way. Like I talked to Ruben Gámez and I realised how little I know about business. It's normal for people to not have, but still you want to try and understand it and what have you. If it's my job to be a programmer, I want to understand it. This is where, honestly, this is where like the SEO and the Google and the AI, I do have a lot of passion about which I didn't realise until we started. I kind of knew because I'm tweeting about it, but it really...

really follows me up.

Paul (12:32.834)
Well, back to search engine stuff. since you bring up SEO, that's another point that I have about why Google search engine is shit. And it's that SEO has ruined it. But not that SEO has ruined it. It's that Google kind of made their search engine reward shitty content. Right. So like if I want to find recipes are the best example of this, I want to look up like if I want to look up like how to scramble an egg of the first

Jack (12:51.436)
Yeah.

Jack (13:01.058)
Yeah.

Paul (13:02.178)
thing will be a recipe on scrambling, which is stupid. There's a recipe for scrambling eggs, but you click it and there's like 200 paragraphs about like my, my, my nan used to scramble eggs for me and blah, blah, blah. And this kind of, it's just like, dude, just tell me how long to cook the eggs for.

Jack (13:06.043)
Ha!

Yeah, a life story.

Mm-hmm.

Jack (13:18.112)
drives me insane and to get actually get the ingredients for a recipe you've got to scroll down or they'll talk about something and they'll actually list out partial instructions and then you'll start following them and then you realize that there's actually more to it that was in the actual recipe down below. I hate it. I honestly have given up with online recipes for the most part.

Paul (13:19.5)
Yeah.

Paul (13:36.322)
Yep. But so that and that's kind of enigmatic of the entire thing where you look for something if you actually see results, it's these SEOified bits of content that isn't really it's not for us. It's not for humans. It's for fucking search engines and it's garbage and it makes it worse. And so I guess one of the main points is and I kind of glossed over this in the beginning, except when kind of as you're bringing stuff up, I'm remembering other points about why Google search sucks.

Jack (14:03.832)
Mm-hmm.

Paul (14:05.944)
So the I guess the main reason is that have you noticed that it's almost entirely ads now like it used to be you want to get to the top of the search results the top of the search results you have to scroll below the full to see. And so it's like four or five ads to start I think they've grown by like 30 or 40 % since 2021 as well like the size of the ads and they also they used to be highly I don't know if you remember this this is from years ago you might not have been born.

Jack (14:13.771)
yeah.

Jack (14:19.053)
Yeah.

Paul (14:35.938)
But Google used to highlight ads in bright yellow. It was yellow. And you're like, OK, I don't even need to read any words. The yellows highlights are ads. Everything else is not. Now it just says a little word like sponsored or something like that.

Jack (14:38.535)
I was born. I remember this. Yes, I was alive.

Jack (14:50.742)
Yeah. Yeah, this is getting worse and worse. I mean, they need to make money. understand that. I mean, they're an advertising company, right? But it does it completely ruins the experience that search engine. I mean, the search engine, you know, never necessarily was trustworthy. But you know, there was you did hope that the right stuff got surfaced. I mean, my I'll be honest with you, my main lens is Stack Overflow. So where I'm leaning nowadays,

Paul (14:59.373)
Yeah.

Paul (15:03.191)
you

Paul (15:08.686)
Mm-hmm.

Paul (15:16.717)
Yeah.

Jack (15:18.454)
I'm leaning more towards source materials. I tweeted this the other day, sorry, Xed it the other day. Aaron Francis a while back was talking about how much he likes to read the manual of things. He's going down, say he wants to learn more about Postgres, he will literally read the entire manual from start to finish. And I'm it sounds insane, but this is the source material, right? Like how is there going to be a better way? You can go into Stack Overflow and learn stuff and learn how people use it, absolutely. But you go to the manual. And I remember,

Paul (15:39.469)
Yeah.

Jack (15:48.118)
I found the same with Single Store. I've read their documentation start to finish multiple times. And what will happen is each time I will find new stuff or have new perspectives on it. so, and same with health stuff, you can actually go deeper into some of the literature, for example. So if I want to Google, if I was to Google about fish oil right now, for example, I guarantee they'd be like, you don't want to have more than whatever a day, but you go into the literature and you'll find something where...

higher amount supported something. Another example is CLL. We talked about this CLL with the chronic was it chronic lymphatic leukemia, I forget the specific thing. You're like, Glenn Saban, you should I worked with this guy, in the past, you should read his book, I read it. And there's a paper behind it. This guy reverses something that is irreversible. that's interesting.

Paul (16:35.894)
Yeah, he was given a couple months to live like 20 years ago.

Jack (16:39.566)
Yeah. So that story is interesting, but with loads of other things and I just go a bit deeper. And I mean, the same thing, like, I don't want to get too into health on this episode, but like things like cholesterol, you eat a bunch of bacon, you're going to get high cholesterol. This guy goes and eats his Nick Norwitz. He goes and eats like 2000 strips of bacon and he lowers his cholesterol. And sorry, he's the guy that, I think I don't know if it was 2000, but it was a lot. This is the guy that compared Oreos to statins.

Paul (17:01.24)
Yeah.

Paul (17:08.942)
Yeah

Jack (17:09.352)
like Oreos reign supreme, but this is what I mean, like doing your own work and doing your own research, whether that be, reading books or taking in more sources, you joked about quote unquote the science that has become a meme at this point. And for me, it's not political. know it can be political with the view and like left wing want this way. I mean, honestly, I view that politicians are all in bed together with that stuff. So to me, it's just, for me, it's just a meme. And it's like, no, I have a duty.

Paul (17:26.04)
Yeah.

Jack (17:39.434)
my family when we were regards to health to do my own research. You know, and it's no good saying, but I'm not, I haven't got a degree in this or, you you're not a doctor. It's like, okay, but you can't think for yourself. But how do you think doctors became doctors? You talk to a scientist and you start talking about a concept. They're like, show me the papers or, you know, whatever else. But then it's funny with the papers, you got to make sure that there was just something that came out in

Paul (17:56.237)
Yeah.

Jack (18:09.354)
from NIH funding billions of dollars, I forget the name of the of the person, but they've just found out that he's been modifying a bunch of his research.

Paul (18:18.357)
the Alzheimer's guy?

Jack (18:20.278)
I forget who it was, but so even then you've got to be careful. But then that's the thing in science. You never use just one paper to support an idea. You want to have a diversity of sources. So any who I'm just more and more, we're talking about search engines and how they're getting worse and worse. I am really doubling down on going to the source because I can Google something and I'll get results that I know from I'm not an expert on loads of things, but there are certain topics that I am.

Paul (18:23.491)
Yeah.

Paul (18:38.851)
Yeah.

Jack (18:48.942)
pretty good with. I'm pretty good with light environment. And I can, I won't find on Google some of the stuff that I know to be true from updated literature, for example. And they will rank centralized medical sites above, you know, perhaps there's a bunch of there's physicians or scientists that are doing research that has come out maybe a few months ago, that's not going to be first on Google, you really have to dive in and find it yourself.

Paul (18:58.648)
Mm-hmm.

Paul (19:13.29)
No, it's going to be like WebMD and Livestrong and those garbage sites, which are just SEO buckets, basically.

Jack (19:18.294)
Yeah, WebMD you can at least get access. That's the NIH one, one, least you can get access to all of the papers from all of the journals. think even when it's not like, I'm pretty sure WebMD you can get access to stuff. I don't know about the rankings, you have also Paul, you have to know what you're looking for. Like you have to add in the keywords else they'll be like, here you go. Like this, this is a, whatever this is surface for whatever reason. But so yeah, I notice it on health, but honestly with programming, I do still think it's good.

I, cause they do default to things like stack overflow, which is good because it's real people with real problems. Seeing a problem and seeing the different ideas on how to solve it and the whys and the debates around the limitations. That is really, really good. And that helps you learn. So yeah, I'm pro learning on that. And Google's going down the drain.

Paul (20:04.141)
Yeah.

Paul (20:10.422)
Yeah, Ike one of the other one of the things you mentioned brings up another point that I have about why Google search sucks and the I guess it's the thinking that you like I guess I guess I didn't really think about this but the the type of results that are shown to you can affect your opinion on something like we were talking about health so it could be like, I hurt my knee I'm going to look up like why my knee hurts.

And those results that you see could affect how you go about trying to like fix the problem. And so there's a researcher called, think his name's Epstein, Dr. Robert Epstein, who's been studying Google. Like he's, he's the guy watching big tech, watching us, him and his team. And so what he found is, and he kind of, cause you know how, if you search for something in Google and I search for something in Google,

Jack (21:00.076)
Mm-hmm. Yep.

Paul (21:08.248)
The results are possibly going to be different. we're in different countries, they're definitely going to be different, but they can be different because Google knows so much about us because they're like a data mining for advertising company. They can affect and he's proven this with like huge data sets that Google can affect how people vote in democracies, which is completely fucked up. Yeah.

Jack (21:18.766)
Mm-hmm.

Jack (21:30.594)
of course. Yeah, absolutely. Well, no, that's everywhere. No, but then that's, that is true. But I mean, look at X, X is now, X used to be very left leaning. And so I would see kind of quite radical left views appear on my timeline a lot. Now I'm seeing both radical left and radical right on my timeline. And like, yeah, well, honestly, exactly. And

Paul (21:50.518)
Yeah, just trying to roll you up to stay on the platform.

Yeah.

Jack (21:56.438)
You know, I think any platform, I think that now X is in the hands of Musk, think Musk has actually endorsed Trump, I believe. So we're going to probably, I would imagine that he would use that power to surface it. I don't know this by the way, but that would be my guess. and then Google is, Google's more left leaning is my perspective. So I'd imagine that they would want to suppress anything that isn't in line with their beliefs. Right? So whoever owns the platform.

Paul (22:10.424)
Yeah.

Paul (22:19.778)
Yeah, I mean, Elon was even tweeting about this when the first, it's so weird to say the first assassination attempt on Trump. Yeah, because he I think he he looked it up in America, the assassination attempt and it showed stuff for like, Kamala Harris. And like, I looked it up and I found stuff about the Trump assassination because I'm in Canada and it doesn't like it doesn't fucking matter to me. Like I can't affect the vote in another country. So

Jack (22:29.416)
there was enough one, wasn't there? Yeah.

Jack (22:39.213)
Right, okay.

Jack (22:48.404)
this is, I love it when people, this is not becoming a political episode, but I love it when people in Canada get rolled up over US politics. You just ask them, so who are you going to vote for? The people we do, we do this a lot. mean, I understand Americans, you know, arguing and debating. I really do because that they can all vote and they can make a difference. And I get that. But like when the Canadians or the Brits argue and they get all uppity about it, it just, makes me laugh. It really does.

Paul (22:55.316)
Yeah.

Paul (23:00.077)
Yeah.

Paul (23:13.92)
Yeah, my God, my life will be over if somebody becomes president. It's like, I don't even know you live there because you don't. but yeah, so Google can Google can shift opinions and voting preferences of undecided voters, apparently up to 20 % or more, which is a lot like that's. Yeah.

Jack (23:19.884)
Yeah, no it's

Jack (23:33.678)
All platforms can, I mean, yeah, that, I don't know, and I take the point. mean, all platforms can. Facebook did, Facebook admitted that they did. I'm a, yes, that was it. Yeah, I'm honest. I'm away from a lot of the news. I've got a new thing on Twitter that completely filters everything. But I remember him writing the letter. I think that was very strategic. I think he's done that strategically because he worries that Trump might win and then be like, you, Mark Zuckerberg. So he's like, has to, he has to sort of balance it. And it's quite funny seeing that.

Paul (23:38.894)
100%. Well, yeah, Zuckerberg wrote that apology letter.

Paul (23:50.53)
Yeah.

Paul (23:57.795)
Yep.

Paul (24:02.669)
Yeah.

Jack (24:02.914)
But yeah, mean, SEO can influence your thoughts on things. you know, AI can do the same thing. You search for certain things and AI is when people search for like us, you can find information about our company and everything else. is stuff in AI. Listen, yeah, AI is probably going to be the future of search. Perplexity can be pretty accurate. I am pretty impressed with perplexity. I just think we have to be honest about the limitations because the frustration I get is not

Paul (24:30.423)
you

Jack (24:32.48)
is not objectively right. Because the frustration I get is from, guess I'd say, I don't want to call them idiots, more and more on is just as bad. But as a group of people saying AI, your job is at risk now, just being grandiose and overhyping it. And then me going, wow, this thing must be really good. You know, I'm digesting this. And then being like, what the hell are they on about? And I'm annoyed.

Paul (24:54.658)
Yeah.

Jack (25:01.09)
because I had that belief, but it was my fault for believing it in the first place, you know, but then it's false promises.

Paul (25:04.96)
Yep. mean, back in the day, had a, because people kept Googling Paul Jarvis Net Worth. So I made a video of me like fanning out my, I think I just sold something on Craigslist or something. So was standing in front of like a 1980s escort car. I'll clarify there with like a very old beater car, fanning out money. And I posted that.

Jack (25:12.683)
I remember this.

Jack (25:26.38)
You, yeah.

Jack (25:30.987)
seen this.

Paul (25:33.25)
And that got into Google search results. And it's just like, that's not real. That's not like accurate.

Jack (25:37.518)
I mean, all the net worth stuff is junk as well. Just I'm thinking about all the celebrity stuff. Yeah, I mean, what's your prediction? I mean, let's get into predictions. Why is it still popular? It's popular because people are using it as what we've always used. What do we use instead? And what do you think the future is? And I think perplexity AI is a big one. I think Google may still remain relevant, but we know the limitations. I'm honestly not sure. What do you think?

Paul (25:40.781)
Yeah.

Paul (25:52.61)
Yeah.

Paul (26:06.37)
I mean, I don't so far AI has proven untrustworthy in so many ways. So I mean, do I think it's moving in that direction? Yeah, of course. I'm not blind to that. I don't think it's necessarily the best idea. think like you said, our ability to think for ourselves and reason is literally what science is. So when people say you got to trust the science, that's not actually how science works. Science isn't just listening to a mandate from a upon high.

Jack (26:33.932)
Yeah, absolutely.

Paul (26:34.86)
doing work, showing your work, arguing different ideas and coming up with different hypotheses. Like that's science, right? Like it's messy. It's super messy and that's how it works. And that's it's always worked. I don't know. I think we're getting stupider. Honestly, like I think we're going to end up like that fucking movie Idiocracy. Like it is pretty good. So I haven't watched it for ages and ages, but.

Jack (26:38.709)
Absolutely.

Jack (26:52.622)
Okay, people keep telling me to watch that, is it?

Paul (26:59.468)
It's just like how everybody refers to the matrix or like V for vendetta all the time when we're talking about like authoritarian and stuff like there's just some movies that kind of encapsulate an idea that doesn't leave like meme status in our brains. So. Yeah.

Jack (27:16.674)
I mean, TikTok is like the kids on TikTok, just some of the stuff is just so dumb. And I just think, yeah, I'm not against entertainment. I'm really not. But when that's your life and, you know, TikTok in Snapchat, I'm trying to get turning into an old man. Yeah, exactly. But no, are we getting dumber? I think, you know, a good job, but then a good way, you know, generations could say this all the time. You know, the kids.

Paul (27:19.818)
you

Paul (27:33.931)
Get off my lawn.

Jack (27:43.982)
Think about the 60s when everyone was doing LSD and mushrooms. You know, like the kids were dumber and then Steve Jobs came out of that. I think there's always going to be a segment of society that's moving us forward. But on average, I don't know. And I don't, I don't think, I don't know, dude.

Paul (27:47.235)
Yeah.

Paul (28:01.87)
Yeah, I guess I trust Google results as much as I trust mainstream media. Like, they might get it right sometimes, but that's not their objective. Like Google's objective isn't to provide us with the truth. Google's objective is to make as much advertising dollars as possible from us looking stuff up. Because search so they I think what are they Q2 of 2024, they made something like $85 billion in Q2 of this year.

Jack (28:06.732)
Yeah, no, for sure. That's about right.

Jack (28:21.228)
Yeah.

Paul (28:32.308)
And 75 % of that came from ad revenue across all their properties, but 75 % of that came from ad revenue just from Google search. So Google search is still their cash cow. Google search is still like their main thing that drives revenue for them. So they want people to use it and they want to put as many ads as possible. And they don't, it doesn't matter what the truth is in search results because they just want to make money. That's kind of how it works.

Jack (28:48.086)
Know the incentives.

Jack (28:58.262)
now you're talking. Okay. So truth also being the adverts suppressing the quote unquote real results and the adverts give you the impression that this is, the adverts give the impression this is really good because it's ranked higher. Understanding. Yeah.

Paul (29:03.981)
Yeah.

Paul (29:08.907)
Yeah, if you look up like best office chair, you're going to find the things that are spending the most money, not the things that have the highest reviews from humans.

Jack (29:15.062)
No, that's good. The word truth. Yeah, I'm still like on the politics stuff, but no, the concept of the truth and what is the best office chair is now diluted. And the way we can, we can counter that is by understanding the incentives, right? Once we understand the incentives, we can understand why something is being done this way. And what have you, but that's requires some, a bit of research or listening to the Above Board podcast where we talk about Q2 revenue.

Paul (29:19.278)
Yeah

Paul (29:28.749)
Yes.

Paul (29:38.702)
Exactly trust us. are the science. Alright dude, I think that was hopefully good. Thank you everybody for listening to the end. Unless you have anything else, I guess we can call it. Alright, cheers everybody.

Jack (29:43.053)
man.