Human society is a large-scale game of biases

There is a blogger on X (my X ID: larry_lawliet) who posted an AI-generated Monet-style water lily painting, added the "made in AI" tag on X, and then asked everyone what the difference was between this and a real Monet. Then the comments started saying that the painting had no soul, dull colors, lacked texture, looked obviously AI, unnatural reflections, lacked depth, like a high school level work, and so on. Then the blogger revealed that this was actually a genuine Monet original. This tweet has nearly 7M views, and the scene became quite awkward—some secretly deleted their comments, others tried to save face. What does this large-scale social psychology experiment reflect? At least two conclusions: First, people’s evaluations of many things are based on labels, stereotypes, preconceived notions, and anchoring effects, rather than any objective truth they believe they have. Second, once people reach a conclusion, they will firmly defend it. Although the brain is not good at objectivity, it is very good at patching subjective preferences that have already been concluded, finding reasons, making excuses, and post-hoc explanations—such as claiming something is bad, or someone is bad, because of xxx; or that the issue isn’t mine but society’s, and so on. Why is this? Because the human brain’s computational power is too limited to fully infer the truth based on Bayes’ theorem; such calculations are too complex. It can only assume some partial evidence as 100% true first, then compare it with existing mental material, seeing which situation or object it resembles more, immediately labeling it, forming a rough conclusion, and then allocating limited resources for subsequent reasoning. Some people even jump to conclusions based on just one label, then use all their computational resources for subsequent explanations to achieve self-consistency—like, if it’s labeled AI, then the premise of “is it AI” is no longer considered, and they directly criticize “AI is worse,” with the brain’s computational power focused on finding “what’s wrong.” Why can’t we think more slowly and accurately? Because survival often requires us to “react quickly,” leaving no time for slow thinking. So, the less computational power one has, the more they need to rely on preconceived notions to reach conclusions as fast as others. Such people are more prone to biases—you can observe this around you: the less intelligent, the more biased, and these people may also speak faster, appear quick-witted on the spot, but actually make more mistakes. Conversely, people like Elon Musk or Steve Jobs tend to react very slowly; they try to reason from first principles, starting from the very first step, avoiding conventional wisdom’s influence, so they seem slow and deliberate, thinking longer because they carefully eliminate each assumption along every decision branch, not taking anything for granted. Do you now understand why there’s a “familiarity effect” in stock investing? For example, if you buy Moutai, you tend to defend Moutai; if you buy Bitcoin, you tend to defend Bitcoin—essentially, all decisions carry biases, just to varying degrees. And these biases are also influenced by group effects: if more people around you say something is good, you’ll think it’s better (if you’re in a Moutai group, you’ll always think Moutai is good; in a Bitcoin group, you’ll always think Bitcoin is good). Because so many people say it’s good, these “evidence” have already made you accept its goodness as true, without further thought. The only thing left to consider is when to buy a little more. It’s precisely because of this characteristic that financial markets inevitably polarize—always irrational, either extremely pessimistic or extremely optimistic. The more participants, the more polarized it becomes, because the more people say it, the lazier people are to think about right or wrong. But if you really want to do well in investing, you can only think slowly, treat most people’s “opinions” as garbage, ignore the voices in your social circle, ignore institutional target prices, ignore stock analysts’ analyses, and use engineering thinking—break down the business into the smallest units: What problem does this business solve? Is it solving it in the best way? How are others solving it? Where are its barriers? How do I value it, and why is this valuation correct? Will others recognize its value in the future, and why? Things like that. Don’t be afraid of trouble—ask yourself “why” at every step, turn investing into an engineering process rather than gambling. #TradFi交易分享挑战