AI is a hot topic for regulators. Regulators and most companies that offer purported AI are clueless about what constitutes actual AI. Way too many charlatans are out there calling a simple trend-following program AI. They create some rules, and the program just follows what they created. That is NOT Artificial Intelligence.
The SEC is all over this and has already been targeting purported AI companies that call their programs AI when they have a simple buy-sell analogy that may trade-off for an Elliot Wave or Stochastic with inherent BIAS created by their predetermined rule. Any system that has an inherent BIAS is not actually AI.
Yes, Bill Gates predicts that artificial intelligence will transform the world in just five years. The International Monetary Fund predicts that the rise of AI could affect about 40% of jobs around the world. When I went to engineering school, we had to learn both programming and hardware. Back then, StarTrek was on TV and that was the inspiration of everyone in the industry to create a computer that was capable of understanding and running the ship. Even Steve Jobs’s inspiration behind Apple and the iPad came from the visions we had from StarTrek.
I worked with Dragon System back in the eighties when it was hardware you put into a slot in an IBM XT. It would allow the computer to talk. My daughter was fascinated by it. I wrote a program just to be able to hold a conversation with her and taught it how to be a politician. If it ventured into an area it did not know, it would just change the subject. I still remember she came home from school one day, and I had the computer apart, and she began crying that I had killed it. I used my kids to teach me how to write natural language so it would understand the words in a conversation. The good old days.
There is a lot of misguided hope surrounding Artificial Intelligence and Machine Learning. All we need to do is look objectively at IBM project WATSON. Just over a decade ago, artificial intelligence (AI) made one of its showier forays into the public’s consciousness when IBM’s Watson computer appeared on the American quiz show Jeopardy. Watson’s debut performance against two of the show’s most successful contestants was televised to a national viewership across three evenings. In the end, the machine triumphed comfortably.
This was NOT actually AI. It is what I call a look-up program. Even ChatGPT is not capable of actually achieving independent thought. They can understand and go fetch the answer faster than a human. The people I knew back then thought Watson would one day cure cancer. None of that was possible because they fooled themselves about AI.
Most of AI’s dangers are caused by its failure to comprehend what it is capable of doing—especially in trading. This all stems from the distorted idea that our brains are supercomputers and there is no God, for our consciousness is simply created by throwing in a bunch of data, shaking well, and out comes a person. Thus, the thrust to mimic the brain led to the creation of neural nets. But that effort also failed in actually creating original thought.
I bought my first Roman coin when I was perhaps 10 to 13 years old for $10. It was in the 3rd century when Rome debased its currency. The coin was from Emperor Aurelian (270-275AD). I noticed that what once was silver coins was bronze, but they used a chemical process to bring 20% silver to the surface to make them look like silver coins when first issued. I saw the same thing in our coins in 1965, issuing clad coinage with a copper center and nickel surface, replacing the last silver coins of 1964. We were doing the same as the Romans.
I was fascinated with the parallel and realized that if I collected the coinage annually, I could see just how fast Rome fell. We all knew Rome fell, but nobody even determined how fast Rome fell. I used the coinage to plot that out. Many people have copied that chart, but you could not make it without testing the coinage.
I also wrote about when I was in 9th grade; my history teacher brought in an old 1937 black-and-white movie named Toast of New York. It was about Jim Fisk and his attempt to corner the gold market that created the Panic of 1869. In that movie, he is looking at the ticker tape quoting gold just hit $162. Because I had a part-time job in a coin/bullion store, I knew the price of gold was $35. Suddenly, what I was being taught in school was wrong. Nothing was linear. It was always a business cycle, for how else could gold be $162 in 1869 and $35 postwar up to 1971?
First of all, this idea that our brains are supercomputers and consciousness is simply a matter of throwing a bunch of data in and out comes consciousness is totally nonsense. A baby knows nothing when born but displays consciousness. My dogs are self-aware. They bark at another dog but put a mirror down, and they know it is themselves. The little one, Josephine, was sick and lying down. I gave Napoleon a treat, and she took it over and gave it to Josephine. I was stunned. They, too, have a consciousness that exists, and it did not come from throwing in a bunch of data.
Any program based on this idea of Machine Language and somehow it will figure out how to trade all by itself is totally absurd. IBM thought Watson could discover the cure for cancer – it did not. As was reported:
“But three years after IBM began selling Watson to recommend the best cancer treatments to doctors around the world, a STAT investigation has found that the supercomputer isn’t living up to the lofty expectations IBM created for it. It is still struggling with the basic step of learning about different forms of cancer. “
Why Machine Learning Has Failed
The entire premise that there is no God and consciousness is achieved by merely throwing in a bunch of data and shaking well has proven to be absolute NONSENSE. A machine learning program cannot teach itself to be a doctor any more than it can become a trader. There is something a lot more to this thing we call consciousness. It will NEVER simply emerge from the data – PERIOD!
You cannot create a machine-learning program and expect it to teach you how to trade, cure cancer, or drive a car. I have raced cars. I have driven almost every sort, even a Formula One—NOT professionally in a Formula One. When you are driving, you have to look at everyone around you. You look for the slightest move, which indicates what that driver is thinking. This is that undefinable gut feeling. You cannot code this, and I have been coding AI since the 1970s.
This is why a self-driving car with AI will not really work. It was a grand theory, but there is no possible way you can expect AI to make intuitive judgment decisions – a gut feeling. This is the problem with expecting that AI will replace humans, where it requires a gut feeling in any endeavor. Penicillin was discovered by mistake. It was not a conscious decision to try to create it. The Economic Confidence Model I also bumped into. I never expected a business cycle that would ever be accurate to the day. I did not believe such a thing was possible. The hope was that AI would then discover things in that fashion by bumping into them – where it did not require original thought.
Therein lies the problem. This expectation that AI can replace human judgment is just fiction. I can mimic emotions on a computer. If you use insulting words, it can hurl an insult back at you. I cannot create actual emotion, nor can I create judgment from a Sixth Sense. This idea that you create a black box, throw in a bunch of data, and out comes an artificial person is absurd.
Someone sent this in when they asked ChapGPT about the difference between it and Socrates. It can look that up and put out the information. It cannot trade any more than it could drive a car. AI will never achieve that human judgment.
To create Socrates, I realized that it could not be a neural net nor a black box where you hope, like IBM, it will figure it all out and somehow emerge as the best trader in history. I taught Socrates how I would look at the world as an international hedge fund manager, comparing everything and looking at the capital flows and COUNTRY RISK. You cannot forecast gold in isolation any more than the stock market. EVERYTHING is connected. As I have often said, the US was bankrupt in 1896, and JP Morgan arranged a $100 million gold loan to bail out the country. Without World Wars I & II, the US would never have become the world’s financial capital. Obviously, you cannot forecast the US share market by just looking at the Fed.
While I taught Socrates how to analyze, I created no hard rules like interest rates up and stocks down. Such market beliefs are in themselves fiction. The Fed was raising rates when Trump came to power, and the market rallied, but they called it the Trump Rally. The Romans used olive oil for heat and light. That was replaced with whale oil, which was then replaced by crude oil. Justinian I (527-565AD) issued the first Clean Air Act in 535AD. He proclaimed the importance of clean air as a birthright. “By the law of nature these things are common to mankind—the air, running water, the sea.” Things evolve, so there could not be any hard rule, such as crude oil always being energy. Socrates had to follow the markets, but it had to have a database of centuries – not weeks or years.
I created no hard relationship rules because relationships ALWAYS change.
