What I Learned Losing a Million Dollars – A Modern Fairy Tale About Gambling in Business

I don’t read business books much anymore. I used to. Obsessively. At one point, reading felt like a compulsion—an intellectual sugar rush I couldn’t resist. These days, I prefer something sharper: peer-reviewed science, niche blogs, curated newsletters, specialized courses, and a healthy dose of GPT-fueled learning. Less time consuming, more frictionless, and far more adaptable to what I’m actually trying to do—learn and build.

But sometimes a book sneaks through the firewall.

What I Learned Losing a Million Dollars, by Jim Paul, came recommended by Sabin Gilceava.

This is an easy read - more of a business fable than a textbook. The story is compelling, a tale of gambling your way into (and out of) trading and business. It follows a tried-and-true formula: tell the reader simple but intriguing truths, sprinkle in some elementary insights from psychology and statistics, and package it all in a way that makes the reader feel smart. It’s accessible. It's a modern fairy tale - i.e. it’s about money.

It dances with ethical ambiguity. You keep wondering: is the author reflecting or justifying? It is not about value creation, nor business. It's about money, connections, bluffing, image, cheating, misrepresentation, risk, gambling, trading, speculation, money.

In that context, the introductory references to Edison or Ford are ironic. Please. Those men were engineers; they built things. 

That said, the book serves as a good reminder of foundational, state-of-the-art scientific and educational literature from psychology, economics, and statistics. The application of the five stages of grief (from pain management) to business loss is actually quite interesting. But like many books in this genre, it overstays its welcome, and sometimes exagerates with elementary truth until they become false. There’s a point where you realize you’re reading another 10-page explanation of why having a plan is better than not having a plan. And surely, both experience and research suggest that rigidly following an initial plan is usually a mistake, something the author simply ignores. Likewise, the value of objective over subjective decision-making is a repeated theme, in literature as well as in this book. Even though, surprisingly, it’s contradicted in the book’s very conclusion.

One quote stands out as a neat summary of the entire work:

“Most people who think they are investing are speculating. And most people who think they are speculating are gambling.”

Simple. Sharp. That wraps it up.

The great modern urban myth: no, AI doesn't replace programmers

AI doesn't replace programmers. It's a modern urban myth.

The net is full of stories about how AI/GPT writes full product code, and replaces programmers. 

It's an urban myth.

It sounds cool. But it's wrong.

AI/GPT is great at boring repetitive tasks. Great at searching. SearchGpt replaces successfully most manuals and documentation. It is even much faster and better than Stackoverflow.

It's also useless for solving any original or complex problem. It makes silly mistakes, writes redundant code, and doesn't really "understand" anything.

Yes, I write code, I love AI, we use it extensively, and I use GPT every day. AI is great. Technology is great.

But no. There is no "product" fully written by AI, by amateurs. 

It's a modern urban myth.

AGI and "true reasoning", vs "AI" and "reasoning. How society handles language mis-use and abuse

Obviously, current AI is not really intelligent, doesn't have "true reasoning" or "true understanding". And we may argue (scientifically) that the terminology is wrong. Which it is. 

But this will not change the way people use, and mis-use, words.

What we do, when words are mis-used or abused, is invent new words.

So, we say AGI (General Artificial Intelligence), and "true reasoning", to clarify that "AI" and "reasoning" are improper terms.

AI intoxication, AI poisoning, and Google bombing

AI intoxication is probably even harder than Google bombing, or Wikipedia astroturfing. Which is already incredibly difficult by now.

But it's coming. It's bound to be a thing,

Same as Google bombing, and Wikipedia Astroturfing.

We need a new name for this. AI intoxication? AI poisoning?

N.b. Google bombing is a method to trick Google to return a specific, fake result, to a specific search query. It exploits Google's algorithm that relies mostly on the number and reliability of websites that link to a specific webpage.

Astroturfing is the practice of creating a false appearance of grassroots support for a cause, product, person, or policy. It manipulates public perception by making it seem like a lot of people are behind a particular opinion.

Wikipedia Astroturfing is a method to trick the crowd-sourced editorial process of Wikipedia, which relies on reputation and wide consensus.

Good guys and bad guys: and the relativity of evil

There are 2 kinds of successful people: good guys, and bad guys. 

The good guys build successful, good businesses; solve real problems; pay solid taxes; help and support good people.

The bad guys build successful, bad businesses; create problems to be solved; use government subsidies, avoid taxes; exploit people.

Then, both guys get rich. 

Their values become relative.

They both built empires, use monopolistic practices, use tax shelters, and lobby the same politicians.

What should they do, now?

Should they stop doing business in totalitarian countries? Or adapt their business to e.g. censor their content to some local totalitarian law?

Should they sponsor the campaign or inauguration of an evil president, or risk being politically targeted? (It's very easy to be targeted, as a big company. Monopolistic behavior, taxes, labor law, discrimination, consumer protection, data privacy. There's always something).

So, you are big. You want to get even bigger, and build bigger Krupp cannons, and smarter Hugo Boss uniforms? Then you sponsor the Party.

This is why good people quiet.

This is how bad people win.

This is how good people become bad people.