Humanity, we've got a problem...
by Bill Bonner
Youghal, Ireland - "AI will change the world. Or, so they say. Bloomberg has a whole new section devoted to it. Microsoft says we need to regulate it. So does the head of OpenAI, Sam Altman. He wants the government to make sure everything in the AI space is carefully controlled: “…major governments around the world can create a program to bring together most of the projects that are underway, so that we can agree …that improvements to AI be limited to a certain annual rate.”
Right. Friedrich Hayek, one of the world’s greatest economists, explained that government planning of this sort can never work. The planners lack the necessary detailed, local knowledge; they don’t know what they don’t know, in other words. But that doesn’t stop them.
The Knowledge Problem: And now, some economists claim that AI will solve what they call the “knowledge problem,” making it possible for the feds to regulate every aspect of human life. Imagine you are driving down the road. A voice comes over the AI Driver Assist: “Mr. Jones. You are driving 2.5 miles per hour over the speed limit. Slow down or we will turn off your car.”
Or…you are reading the news. A voice: “The story you have requested has been deemed unreliable by our AI Truth Protection algorithms and therefore is unavailable.” Or…worse. “The polls close in an hour. We remind you that you must vote for the candidate of our…I mean your…choice or your Social Benefits, including your Universal Basic Income, will be terminated.”
A brave new AI-driven world? Maybe. What is AI really doing for us? We ask ChatGPT to explain: "AI tools like myself and Google's Bard can potentially reduce economic input costs… automate tasks… improve the efficiency of production processes… help businesses to make better decisions, which can lead to increased profits.
AI has the potential to be a major driver of economic growth. In fact, some experts believe that AI could be the biggest reduction in economic input costs since the discovery of oil and the invention of the steam engine and the internal combustion engine."
Old World vs New World: If AI is so great, how to get in on it? ChatGPT (AI) has an answer for that too: "Nvidia is a major player in the AI market, and its chips are used in a wide variety of AI applications. As a result, Nvidia is well-positioned to benefit from the growth of the AI market. Some experts believe that Nvidia could be the most important resource company since Standard Oil."
Oil provided energy to move things – machines, tractors, factories. You can test the importance of it yourself. Just stop your car on a hill. Then, try to push it up by hand. Ultimately, energy conserves time. It takes a long time to push a car up a hill – if you can do it at all. It takes much less time with the help of 70-octane gasoline. Oil – from Standard Oil and others – made it possible to haul, hammer, and heave more stuff, freeing time to study things and invent things which led to even greater productivity (output per hour). That’s why our standards of living are so much higher today than they were 200 years ago.
Oil moves tangible things. What does AI do? It moves words… letters… imagines… information…symbols…As we’ve seen with the Internet Revolution, more ‘information’ did us some good, some harm and a lot in-between. On average, Americans already spend 11 hours a day on some sort of electronic device – phone, tablet, computer. They are saving time. Sending a text is a lot faster than writing an old-fashioned letter and putting it in the mail. But nonsense and entertainment expanded to fill the time available to them. Time was wasted.
GDP growth went down. Information confusion increased. And as we’ve discussed many times, the typical American now has to work substantially more time to afford his wheels and the roof over his head.
AI promises to save even more time. If you want an answer to a factual question, AI can get it to you fast. Let’s test it: Which country produces the most copper? Two seconds later, AI has the answer: "As of my knowledge cutoff in September 2021, the country that produces the most copper is Chile.
We can expect AI to generate decent answers on questions that require some calculation or comparison, again saving time. Which is more likely…getting struck by lightning or dying of Covid? Statistically speaking, the likelihood of getting struck by lightning is relatively low. According to the National Weather Service in the United States, the odds of an individual getting struck by lightning in any given year are approximately 1 in 500,000. The risk can be further reduced by following safety guidelines during thunderstorms.
On the other hand, the risk of dying from COVID-19 is influenced by several factors, including age, overall health, and access to healthcare. The fatality rate varies among different populations and countries. It is important to note that the COVID-19 pandemic has affected millions of people worldwide, leading to significant morbidity and mortality."
“Complex and Interconnected”: Let’s take it a little further…we want to see if AI can save the time we would otherwise spend thinking. We ask: what has been the most important consequence of shifting away from a gold-backed dollar in 1971? ChatGPT replies: "The decision to move away from the gold-backed dollar in 1971, known as the Nixon Shock, had several significant consequences that reshaped the global economy and monetary system. One of the most important consequences was the shift toward a fiat currency system, where the value of money is not directly linked to a physical commodity like gold.
Increased Monetary Flexibility: Moving away from the gold standard provided governments and central banks greater flexibility in managing their monetary policies. They gained the ability to adjust interest rates, control money supply, and implement economic stimulus measures more freely. This newfound flexibility allowed for greater responsiveness to economic conditions and the ability to address financial crises.
Inflationary Pressures: One of the key consequences of abandoning the gold-backed dollar was the increased potential for inflation. With the gold standard, the money supply was tied to the availability of gold reserves. Without this constraint, governments could print more money and increase spending, which, if not managed prudently, could lead to inflationary pressures. This has been a concern for economies that have experienced excessive money creation.
Currency Volatility and Exchange Rates: The shift to fiat currencies introduced increased volatility in exchange rates among different currencies. The absence of a fixed gold peg allowed currency values to fluctuate based on market forces, economic conditions, and government policies. This volatility impacted international trade, investment flows, and global financial stability.
Global Reserve Currency: The U.S. dollar, which had been the primary global reserve currency under the gold standard, continued to hold its dominant position after the shift. However, the move away from the gold-backed dollar eroded some of its intrinsic value and raised questions about its long-term stability as a reserve currency. It also opened the door for the emergence of alternative reserve currencies, such as the euro, yen, and, more recently, the Chinese yuan.
Financialization and Speculation: The shift away from the gold standard contributed to the growth of financial markets and increased financial speculation. With fiat currencies, financial instruments, and derivatives, new opportunities for investment and speculation emerged. This expansion of financial markets brought benefits in terms of capital allocation but also introduced new risks and complexities to the global financial system.
It is important to note that these consequences are complex and interconnected, and their impacts vary across different economies and regions. The shift away from the gold-backed dollar marked a significant turning point in the international monetary system, influencing economic policies, financial markets, and the global economic landscape in subsequent decades."
More Blah-Blah: It’s not a bad answer. AI searches through millions of pages of ‘content.’ It then arranges its findings in a socially acceptable, academically correct way. It seems ‘intelligent,’ in the way that a bright person seems intelligent when he says what you think he ought to say. It has the ‘right answer.’ And it saves us the time of trying to figure it out ourselves. But it’s blah-blah. AI sifts and sorts…it homogenizes and certifies, providing the most ‘mainstream’ ideas, information and opinions – that is, those that are most common and most valued by the rest of the world.
As near as we can tell, it has no source of independent thought – ideas not supported by the mass of information and opinion already available. It merely streamlines access to “common knowledge” and saves time. It should reduce errors too. If A is bigger than C, and C is bigger than B…is B bigger than A? The typical voter found at a political rally will reply with something such as “bigger than A what?” AI will be smarter.
But wait…colleague Chris Mayer says he had to correct an answer given by ChatGPT, whereupon it apologized for its mistake. Likewise, our son Henry asked ChatGPT about Fed policy and got a summary blah-blah answer. He wanted to know if AI could deduce Fed policy – inflate or die? So, he sharpened the question: what’s most important to the Fed, financial stability or the value of the dollar? The machine replied that: The “Fed has a double mandate (a dual mandate) that consists of promoting:
Full employment.
Price stability.
Moderate long-term interest rates.
“The fact that there are three rather than two did not bother ChatGPT,” says Henry. Yes, there are bugs to be worked out.
No doubt, AI will bring change. Much of the world’s literary, clerical and informational output is blah-blah. Most people speak blah-blah. Politicians rarely depart from it. And people seem to like it. These new, smart machines can give them all they want.
Those thousands of customer service workers in India, for example, may have to find new jobs. Professionals, too, may find their burdens somewhat lightened. In law, where following precedent (stare decisis) is the rule, AI ought to be useful. And in medicine, where there is a growing body of knowledge and guesswork available electronically, AI might find an important role.
And perhaps in some great and glorious future, America’s president will be wired for AI…responding immediately to the latest polls…providing the blah most likely to meet the approval of AI-shaped public opinion. He will speak blah-blah…the press will report blah-blah…the schools will teach blah-blah…and AI-assisted goons will keep us in our place.
Who knows? Our brother in law – a Southern Baptist minister – says “At aowa age, the two most important things to know ar where ya ar and who ya ar.’ Maybe AI can help."