This sounds like a very esoteric question, but it's actually a really practical one. The difference between how to find the truth isn't always the same as knowing the truth because everybody has biases. As a former senior research analyst, for many years it was my job to figure out the truth or at least figure out the strength of arguments on why I should be long or short a given stock or real estate investment or why is the market moving on a certain news or not or how a news story that comes out in the world and affects the financial markets, legitimate or not. These are some questions I had to tackle and still tackle every day as part of my career.
And working in the financial markets where your money is on the line, if you were, wrong really hones your senses and your BS detector because you have that accountability driven into it. And so for years of doing this and also years of just reading tons of books about a variety of different topics and also kind of growing up in a more discussion-focused extended family. Also pursing PhD's just teaches you how to research.
If you're going to make bold and assertive statements that are not like Wikipedia level obvious, you should be able to have something to back up your claim, whether it's a hard source, whether it is a strong logic or physical evidence or a combination of the three depending on what the topic is. In our world, technological advancements have increased the fog of war in society and not decreased it. You think with the advent of the internet where every piece of information is at your fingertips or with LLMs like ChatGPT and Grock and Claude and Gemini, I can just type in any query and it should give me the answer. But despite all this easy access of information with a push of a button on the computer, it's actually done the opposite and made truth harder to find for a variety of reasons.
One, you have the hallucination rates of LLMs. They're anywhere from 5 to 30% depending on which topic you are dealing with. And the key really to finding truth for LLMs is you should always ask for the source material for any claim made in a conversation. So say for example, I'm trying to argue why the sky is blue and it gives me some hypothesis that the sky is blue because Apollo in ancient Greek times painted the sky blue. And I'm like, well, that doesn't sound right to me. What are your source materials for that? And if yo have any sort of question of the accuracy of the claim that you're getting the questions answered to, you should ask for source material. Or if you just legitimately don't know the answer, like if you're asking like what is the GDP per capita in Botswana and how has it changed since 1950? And it just throws out some random numbers and you don't know whether they're right or wrong. You say, "Hey, please tell me the source information and how did you find that data?" Then you can read the source material for yourself. If it can't replicate it or if the source material doesn't match, then you know you got a hallucination.
And then you also have the development of Sora, which video evidence used to be, especially when I was a kid, stronger. People always said, look, I don't believe something until I see it on video. A photo can be doctored on Photoshop. Audio can be manipulated, but a video means it's a pretty good chance of being true. But now with Sora and AI video apps, you can create like a 90 second or less video about anything, any person doing anything. As this gets better, the length of times are going to get longer. So now you can't even trust video evidence because a lot of the video evidence could be manufactured. Same with audio or photography or you can have forged documents. So the things that used to be quote unquote hard evidence aren't as hard as they used to be.
And not even going to the political incentives of post-truth, fake news, whatever you'd like to call it, where political candidates on either side of the aisle now prioritize emotional appeal over facts and logic and trying to create a narrative. And by having your audience captured to a narrative, you are more willing to control their actions and decisions. People will change their beliefs or report things selectively to match that narrative, even if that narrative is not objectively true.
And you could have even financial markets or businesses make decisions based on narratives that aren't true.
There's not a single book or media outlet or social media handle to follow or YouTube channel that is going to have all the answers or always be 100% objective and accurate. That is simply not the case. There's just too many incentives, both politically driven and non-politically driven. And there's just the human error or LLM hallucination that will not make any single source a guaranteed true source.
So if you're looking for just some simple answer like a list of five books about how to be able to discern truth, you're not going to find it here. But a few key things. First one thing you have to factor is the god gene or the scientific name is solute carrier family 18 member 2 SLC18 A2. It's also known as the vesicular monoamine transporter 2 vmat 2. And this is kind of known as the God gene because this gene has been proven to be associated with people's sense of spirituality, dopamine and serotonin regulation, and susceptibility to want to find meaning and the openness to metaphysical claims. It really is the propensity to believe, trust, and adopt a narrative of something transcendent beyond the observable. Not all people have this gene expressed in a certain way. And we live in this world, we're in a secular society now. People don't really have devout devotion to anything, whether it's God or anything else. We're beyond that. We've evolved past that. Well, that is the trap because just because you don't really adhere to traditional religion anymore doesn't mean that you don't have this god gene and this instinct to find something transcendent. You're never going to be able to logic your way to 100% into any sort of religious tradition. You have to acknowledge that you have this God gene and this desire for something transcendent.
For religious people, it's easier. It's placed in their religion. But for non-religious people, you have to figure out where your God gene is placed. Some people place it in their sports team in devotion to that. Some people place it in devotion for celebrities or pop culture icons. Some people put it in nationalism or devotion to their political parties. I think that's the most common one in modern west is you replace the devotion of religion, the transcendence of religion with devotion to a political party and a transcendent political ideology.
But it could also apply to sports teams. It can apply to celebrity worship. Can apply to the worship of chasing a certain sensation, whether it's the pursuit of wealth, romantic relationships, or anything that has a greater meaning. People have their religious devotion to a coin. You have the myth of Satoshi Nakamoto. But you have to figure out if you're not somebody who places this kind of god gene into God itself, where do you place it? A good friend of mine puts it into stoicism.
And realize if you have a god gene level devotion to this kind of thing or believe it is some sort of transcendence in something there is no hard evidence for that transcendence you have to realize I cannot be objective about this topic and therefore any conclusion that I make about it will be heavily clouded you either can say hey look I need to place the god gene somewhere that doesn't result in a detriment in my pursuit of truth or you don't try to pursue truth in that given category.
So that's kind of step one. Next, when it comes to step two is the boundary condition method. You got to take any claim and ask logically at what point does the claim break its own logic.
Like an example would be housing prices. You see all these claims like housing prices will always go up faster than inflation. It's always the best investment to invest in real estate in the long run over any other asset class. Housing prices have gone up, say, I don't know, 5% a year or 6% a year the last 20 years. They're going to go up 6 to 10% a year the next 20 years because they're not making any more land.
But is that an actual like hard claim? And what is the basis of that claim? I mean, I think the answer is no. That's not true. I think I posted a thread on here for my opinion on this.
But the greater answer is, let's just say you're right. Let's say if housing goes up double the rate of inflation every year for the next 25 years, just like it did the last 25 years, and it goes up relative to wages as well. What would the world need to look like for that to be true?
You would need to either have a very large homeless population, renormalize multi-generational like three or four generational houses or have the living standards go down to countries that have similar high price income ratios like in Southeast Asia where people live in basically squalor because housing is so unaffordable and live in extremely high density.
When you look at the data on that in the US, that's not the case. The average square footage of housing has consistently risen until the last 5 years. And the population growth is not really warranted to sustain that. It's actually the opposite. It's probably more of a demographic decline that's going to put downward pressure on housing. If housing prices get too high, basically there will be nobody who can afford to live in those houses and there'll just be a bunch of empty houses and homeless camps or everybody living in like third world squalor. And you don't realize that's the argument you're making, but that is effectively the argument you're making.
And it's not just housing. Look at anything, any claim you make and draw its logical conclusion. And when you realize if you take something to its furthest conclusion, it might sound absurd, but if the logic breaks at a certain point, it means that the claim is not inherently true.
The next real stage is when it comes to news sources in general, you have to feel, look and see what are the implicit biases in the news sources, which is obvious one. There's a lot of them have ideological bents. Like I'm not sponsored by ground news or anything, but they do a good job of sorting on the political spectrum certain editorial or editorialish reports. Look at which news reporting agencies have readers that care the most about objective truth.
The answer is usually either industry journals or financial press.
I'm talking about the journalist side, not the editorial side, because even Bloomberg and the Wall Street about the journalist side, not the editorial side, because even Bloomberg and the Wall Street Journal have very slanted editorials that cater to specific ideologies. But the news reporting for something like Bloomberg and the Wall Street Journal are fairly objective. Why are they objective? Why do I trust them other than other mainstream sources of journalism? because they need to be objective because their readers are investors and financial people who need news that actually tells the truth about the situation because they're making investment decisions on it.
And if they realize that the WSJ or Bloomberg has editorial slants that provide some manipulation of the truth in their general reporting, they're going to stop relying on it and therefore they're going to drop their subscriptions and be like, "This is a waste of my time." CNBC and Fox Business are good examples of that. Those are both very heavily editorialized. They're heavily biased by their advertisers as well. As a result, no serious investor takes either CNBC or Fox Business seriously as a source of accurate news.
Maybe the headlines and the stock tickers on the bottom of the screen are correct, but if anybody has them on at their office, they're in mute. And same with Bloomberg TV, too. If they do have on the sound, they know it's just meaningless squawk that is not reliable information.
Industry journals are generally good at this because people have to buy a subscription to that journal want accurate information about their industry to make good business and corporate finance decisions. If it has some editorial slant, you either alienate half your audience due to a partisan bias or you're not going to trust it. I think substacks and particularly paid substacks can be good for those given niches, but it depends what people want. If they're reading that paid substack to validate their own ideological feelings about something, then it's not an accurate source of objective truth.
You
can get objective truth from editorials, but people are not paying to get accurate information. They're paying to see a worldview that validates their own worldview. There is value in that. But in terms of if you're trying to figure out what is more likely to report the truth, you have to look at the incentive of the messenger.
The next real point is this idea of the dog doesn't bark, which comes from Sherlock Holmes, that he found who the murderer was in a case because the dog didn't bark in response to that suspect because the dog knew that suspect.
But when it comes to what this means for your pursuit of truth, you have to look at a controversial story. Say for example, there is story about a flood of 10 million immigrants crossing the Canadian border into America and wreaking havoc in towns in Minnesota. You see a bunch of right leaning news reports about it. Is that accurate? Is that true? You can't know because the right leaning reports who generally have editorial biases against immigration are likely to exaggerate the numbers to make the problem seem worse to inspire people to take more action about this issue.
The thing to look at is not just the report from the right-wing news outlet, but to see how the other side responds to it. So, let's just say for example, the other side responds and says the numbers are actually instead of 10 to 20 million, it's maybe 1 to 2 million people who have crossed the border. And they show statistics for border crossings as evidence for this. And then you think that's the end. No, you'd have to see if the other side has a response to that and shows that those numbers might not be accurate because of X, Y, and Z and has evidence to prove that, then that doesn't prove anything.
So basically if you have each side have a counternarrative that addresses all the objections and the concerns on other side then you don't really know what's entirely true. You can't determine it just on one source reporting against the other.
What you can discern though is when it is more unbalanced. Let's just say you see the same initial right-wing report of 10 million migrants crossing the Canadian border and moving to Minnesota. And the left-leaning people don't respond or they say, "Oh, that's just racist." But they don't really have anything to back those claims or they just ignore it and don't respond to it at all. That is a sign that there is truth there because the dog is not barking.
Like if it was false, the sides that are in the interest of it being false will be the first to report it. This isn't just a left or right thing. This could be like the bull or the bear case for a stock, too. If the bears report something bad about a company and the bulls have no response to it or just a typical saying that's not true with no substantial really claim why it's the case, then the bear might be right or vice versa.
The bulls paint a rosy narrative or the bearer makes a claim and the other side responds to say, "Hey, the numbers are actually this, this, and this." And management shows strong reputations, then the bear is wrong. Since the bear has no response to those reputations, it's basically finding what is the truth, not as what is spoken.
And if there's something with substantial facts that people are chattering about that neither major outlet is willing to discuss, there probably is some greater truth to it, assuming it passes all the other logical tests. A lot of conspiracy theories could fit in this, but they don't pass a lot of the other logical tests that we're going to run in this analysis.
The next thing is the financial market reaction. If something is real news that really matters, the financial market is going to move as a response to it. Let's say for example, there is a new scientific report that eating eggs causes men who eat more than say one egg a day, they have a 20% more chance of going bald. If this was true and got into the mainstream news, eggs manufacturing stocks would go down and react negatively to that because probably a lot of men afraid of losing their hair would not eat the eggs anymore.
Or like another thing, let's just say there's some story about, oh, there's a new war going on in the Middle East and in an oil region and you don't see oil stocks move. It's either that the war is not actually happening or if it is, it's minor and will be resolved quickly because if it was something serious and a real threat, the oil market would react because the oil market is something that reacts the most for geopolitical conflicts in the Middle East.
And it's not the same thing that will react to every news story. But if it's something that's a real news story that's affecting something real in the economy or in society that translates to economic or purchasing decisions, it should within enough time have an effect on the financial markets. You just often don't find the news literally the second it comes out.
So if you see a report about a month ago showing young people no longer buy video game consoles, you would see some price action in the video game stocks that they would be selling off. But if you don't, you know that study is probably not accurate or there's other variables at play that are not being told.
So, I think one of the most accurate ways to get down to the truth of something is is the market reacting with events that aren't necessarily purely tied to a financial instrument. You should see a reaction in the prediction markets. A material reaction in the prediction markets as probably a higher degree of reflecting the veracity of a claim than just reading a headline somewhere.
And then also, you have to look at the source of that headline too. Like an anonymous Twitter feed is a lot different level of credibility than the Wall Street Journal. So, now that we've talked about like the financial markets and the boundary condition and the God gene and the and the dog not barking. Those are four main logical tests for whether a truth claim is true or not. I think it's really helpful to read a lot of books or learn a lot of things in different competing or separate disciplines.
Having an interdisciplinary education will help you because you'll see some logical inconsistencies that are common in one field of study that are not in another. When you see rhetoric shift from one place to another, you'll recognize the logical fallacies or the inconsistencies from your research in other topics. And your knowledge about other things can be used to heighten your BS detector and also be able to understand better the logical conclusions of certain claims.
On your path to really finding the truth or just being able to discern fact from fiction, there needs to be a common sense element integrated into this. You need to see if the claims you are reading or the statistics or data that you're seeing in the news or any fundamentals of businesses, your local community or politics or whatever the thing you're looking up if there's some common sense that is anchored to it.
Like if you're reading all this stuff like yeah that theoretically I don't know log makes some sense and there's some logic to it then maybe it's true but you have to align it with your common sense and perceptions based on observed reality. Like if you draw a conclusion that just simply does not line up with observed reality. There might be an indicator that your previous assumptions are not true or the analysis you are doing is flawed.
However, you can't base it solely on your own experiences because data is not simply plural for your own individual anecdotes and anecdotal evidence is not sufficient evidence because of something what I call and probably others call the bubble effect. There's famous historical examples where people in Manhattan said, "I never knew anybody who voted for Nixon in 72." Even though Nixon had the biggest landslide in modern electoral history at that time, yet the Manhattan Social Circle, nobody knew anybody who voted for him.
Is your common sense based on real common sense? Or is your common sense based off of living in a bubble environment or socializing in a bubble environment with a very similar like-minded cohort or a base of assumptions that really only apply to your region or your subculture that may not apply to the world as a whole.
And then the last one is skin in the game and stake because reputation of journalists is not enough really skin in the game to determine if something is true or not. The most trustworthy information is probably quarterly reports from financial comp from quarterly earnings reports for S&P 500 earnings releases because the CEO and the CFO have to sign an affidavit that what they are saying is accurate, complete, and true.
If it's not, they face criminal liability from the SEC. If your status as a free citizen is on the line for you to tell the truth, you're far more likely to tell the truth.
That doesn't mean there's not fraud. There are examples of securities fraud and accounting fraud in markets. But for all the examples of fraud, there's 99 times more or 100 times more examples of honest and accurate information.
It's not just in finance, like any organization that there are legal and serious reputational penalties. Like you might not not like, oh, he's a doofus because his opinions are stupid. I mean, no. Like, you might not be able to get a job in your field if you're this wrong, or you will lose your election for sure, or your career or reputation is destroyed, you face jail time, or whatever. There's some serious consequence for you to mislead people. That is more likely to be trustworthy than somebody who has no skin in the game for being wrong. Like the example would be somebody's betting $100,000 in the prediction market is probably (
hopefully) more likely to be accurate on their view because they have $100,000 on the line than somebody who is just posting a random anonymous comment on Reddit or Twitter who has nothing on the line because nobody even knows it's them. So that shows you the difference in the magnitude but you have to look at the skin in the game and also it reflects bias like who is funding the research or who is promoting a certain narrative.
If for example you are trying to do research about agism in the workforce and the study is done exclusively by technology companies who are naturally biased to under report corporate agism in the workforce. You have to take that with a grain of salt. Like on the other side let's just look at AARP. If AARP does a study about agism in the workforce, they are a advocacy group and a lobbying group for Americans over the age of 55 years old. So they're going to have a bias to claim there is more agism because that means that the government needs to crack down on this agism in the workforce because they're representing the interests of older people. And so the AP study is likely to overestimate the level of agism and the tech companies funded study is likely to underestimate the role of corporate agism.
And so you have to look at the sources and if you dive deeper into this study the other thing you have to look at is the methodology like how did they collect this study? So say if you see a survey saying women no longer want to get married and it's like well how did they get that conclusion? Oh, I interviewed 10 women and a women's studies graduate program at Berkeley and out of those 10, only one of them wanted to get married. Well, I mean, look at the sample. It's people who have naturally a anti-male, anti-marriage perspective. And you're saying that is representative of all women who don't want to get married. where if the study is 100,000 people across the entire country and not just on college campuses, age groups from 18 to 50, that might be a more accurate representation of whether women still want to get married or not. If you did it inside a single congregation of a very devout parish, that might skew the results even and if those results are anti-marriage, it'd probably be even more accurate because that'd be a group that would be predisposed to being prom.
So, you have to look at the study, the sample size, who is the cohort that is being surveyed, if it's sociological data or if it's scientific data, you have to look at the replicability. There's a real issue for studies called the replicability crisis where a large chunk of studies in psychology and the hard sciences are not replicable which means you can't redo them and get the same results consistently within a certain confidence interval and therefore invalidates the entire study.
A lot of financial studies have this problem because we don't have that many years of financial markets operating the way they do now. The sample sizes are often not big enough to account for the nonbell curve shape of financial returns. The other real thing is in economics. This is why I'm a big critic of Austrian economics. I have a lot of sympathy for the policy solutions of Austrians and that's because they rely on praxiology which is purely drawing things to their logical conclusion and not trying to empirically test any of their claims. If a claim is not testable, it is a somewhat (quasi-)religious belief.
Like the claim that equality is the highest virtue is a religious belief. It is not an accurate claim of truth because there is no way to falsify it or test it. Or let's say on the other side like the free market solves everything. Every social problem can be solved by the free market.
You can believe that if you want but that is not an objective truth claim because how do you falsify that? How do you measure whether a problem is solved because of the free market or not? Free market on the net does a lot of good for society and I generally am sympathetic to the free market personally
but the evidence really that the free market can solve every social problem is not based on any sort of truth. It is a religious belief.
So you have to figure out is the thing that claim is something that can be falsified. If it can't be falsified, there's no real point in trying to take it for gospel or debate on it because you can't draw any sort of conclusion. The other thing too is that if it is something you want to debate, it is a controversial doesn't mean there's no truth. Sometimes maybe you need to be the one who tries to test it if nobody has or you got to find somebody who has tried to test it because any claim that is testable, how many academics and scientists and amateur researchers we have out there is probably being tested somewhere in the world. So you can find it or do it yourself.
But the idea is that you can discern truth, but it takes a large amount of critical thinking and deep analysis that quite frankly most people do not have the time or inclination to do so. Some of these things can be done rather simply. Others take a lot more deep dive. But if you if you but the thing is the good side is even in the age of AI distorting things. It's the same technology that allows for these distortions also allows for greater access to the actual hard facts behind any narrative.
There are other heuristics involved and I probably focused too much on the financial markets but I hope that helps
A friend of mine has a bachelors in English or something, but a masters in statistical modeling (basically how to design studies and account for bias), and she has a PhD in behavioral psychology. She would be a better one to ask.
View attachment IMG_5501.webp
