Pandora's Box
Written by Paul Siluch
December 1st, 2023
In Greek mythology, fire was a gift reserved for the gods. Watching humanity shiver in the cold, the god Prometheus took pity, stole fire from Mt. Olympus, and gave it to the mortals below. Zeus was furious. He chained Prometheus to a rock and had an eagle eat his liver every day, which grew back each night.
Okay, he was immortal, but that is still one heck of an eternal punishment.
Zeus was known to be exceptionally vengeful, so he also sent a small chest containing all the evils of the world to the wife of Prometheus’s brother – a woman named Pandora. Zeus knew Pandora’s curiosity would impel her to open the box and, once she did, out flew sorrow, disease, vice, greed, and madness. Only Hope remained inside when she slammed the lid shut. Had it escaped, all hope for mankind’s future would have been lost.
Pandora’s Box has become a metaphor for something that causes unforeseen problems. Similar allusions are the Law of Unintended Consequences, and “to open a can of worms.”
We may have witnessed the opening of the modern Pandora’s Box this month when the company OpenAI, which we know as the inventor of the ChatGPT artificial intelligence algorithm, fired its CEO, Sam Altman. OpenAI was worth approximately US $80 billion on paper the day before the firing and was literally cut in half a day later when after Mr. Altman was booted out. Over 500 employees demanded he be reinstated, and the board of directors fired instead.
Now that is employee loyalty.
Underneath the drama is a far bigger issue: the advancement of Artificial Intelligence (AI) in a responsible and safe manner.
Fear of Change
Humans both seek change and are terrified of it. Technology accelerates these fears and raises the stakes with each new invention.
The Luddites were a group of loom weavers who saw their jobs disappearing. They destroyed every steam-powered loom they could find and set back the pace of modernization. This worked for a few years, but the army and the moneyed elite steamrolled the Luddites soon after.
When books and leisure reading became widespread, we were told they led to laziness and brain rot. They didn’t, but the same ills were transferred to television in the 1950s. Ray Bradbury’s novel “Fahrenheit 451” described a television on every wall in our homes and the subsequent burning of all books.
When we split the atom and made the first atomic bomb, movies such as “On the Beach” about global extinction and “Dr. Strangelove” about a madman in charge of the A-bomb followed.
Yes, uranium was used for bombs, but far more has been directed toward nuclear power - the cleanest form of electricity on the planet. But bombs and extinction make for better drama. The risk of nuclear war is real, yet the benefit of unlimited energy is a far greater benefit.
And so, we come to artificial intelligence and the fear of computers taking over. We’ve been afraid of this ever since the first computers were invented.
Science fiction has explored this theme in gory detail.
Harlan Ellison’s “I Have No Mouth and I Must Scream” was an award-winning story about rival U.S. and Soviet computers that realized a better victory was to team up and enslave humanity.
And of course, the Terminator movies introduced us to the evil Skynet computer overlords and their cyborg soldiers.
In geological terms, human intelligence has evolved at a lightning pace, from knuckle-dragging tree climbers to gene-editing scientists in only a million years. Some biologists say we are evolving faster today, even as we can’t see it happening.
But computers evolve at a much faster pace. Moore’s Law states that computing power doubles every 18-24 months. If a computer brain were to become as intelligent as we are today, it could be twice as intelligent as a human brain in just two years, and then double that again in two more. It wouldn’t be long before we once again appeared as knuckle-dragging apes compared to a machine intelligence.
The more you improve yourself, the faster you improve yourself. Within a few cycles, you develop the intelligence of a God.
- Unchartedterritories.com
AI Armageddon is not a given, of course. Automated looms, TV, nuclear fission, and the internet have only benefitted mankind. Machine intelligence is already greatly speeding up how we work, how we think, and what we know. The dark side is losing control of this advancement. If we made a machine whose sole mission was to make paper clips, for example, how long before an even smarter machine decided it could make more if there were fewer people getting in the way?
While the odds of a runaway machine intelligence going rogue are assessed to be low, they are not zero odds. Which is the reason OpenAI was created in the first place.
OpenAI was founded as a non-profit society in 2015 and was not a company at all. OpenAI began with such founders as Elon Musk based on a shared concerned about the ethical and measured advancement of artificial intelligence. We will only get one chance to get this right, they reasoned.
It was formed with the following mission:
“To build artificial general intelligence (AGI) that is safe and benefits all of humanity.”
Underneath the non-profit society, a corporation was formed that is designed to make money. This is now owned 49% by Microsoft as well as by employees and the OpenAI society above it.
Thanks to the progress of the corporation, ChatGPT has evolved to something far more powerful than what was envisioned in 2015. Yet even this was not fast enough for Sam Altman, the CEO. He wanted money to make special chips and to hire more engineers to push the technology even faster. Some reports claim his team made a transformative breakthrough just last month that could enable the next step to true machine intelligence.
If the horse (the OpenAI society) is supposed to lead the cart (the ChatGPT corporation), the cart decided it no longer wanted to follow a slow horse. OpenAI was concerned enough to terminate its CEO’s employment.
Microsoft has the most to gain or lose from the success of OpenAI’s lead product, ChatGPT. The shares fell sharply on the news of the firing, then rebounded when Microsoft said it would hire Sam Altman and anyone who wanted to join him.
Money talks. ChatGPT has far more allies and incentives than the cautious board of directors at OpenAI does, so Sam Altman is now back as CEO and a new board has been appointed.
The billions in potential profits from advancing AI drowned out the cautionary voices calling to slow it down. AI development is likely to move even faster than before, and Microsoft will have the biggest seat at the table.
Why AI is (Likely) Going to Be Fine
When Robert Oppenheimer developed the atomic bomb at the famed Manhattan Project, he calculated there was a remote possibility that the atomic blast might start a chain reaction that could set fire to the atmosphere and destroy the whole world. It was a remote chance, but one he had to carefully consider. Offsetting this was the race between the Allies and Germany to split the atom first.
Yes, there were risks, but since nuclear fission was inevitable, it would be better if the good guys (us) got there first. The Los Alamos explosion did not ignite the atmosphere, much to Dr. Oppenheimer’s relief.
And the world’s.
The race to develop artificial intelligence parallels what happened in 1945. Both the U.S. and China are competing to lead AI, which will determine the winner in the next generation of technology. AI is already being used in everything from spellcheck (the word most likely to come next is suggested as you type) to daily calendars and web queries. What comes after is fully autonomous driving, improved medical diagnosis, and the development of materials that don’t yet exist in nature.
Human-machine productivity is about to go through the roof. Every category of efficiency, from profit margins to customer satisfaction, is expected to improve with the use of AI:
Source: Capital Group
A recent study by the European Central Bank found that artificial intelligence is creating, not destroying jobs, especially for the young and highly skilled.
That is, as long as artificial intelligence doesn’t go rogue.
Many people working on AI are concerned about this and are dedicated to making sure developments are friendly to humans. In fact, the philosophy goes back over 80 years to science fiction writer Isaac Asimov, who proposed the Three Laws of Robotics:
The First Law: A robot may not injure a human being or, through inaction, allow a human being to come to harm.
The Second Law: A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
The Third Law: A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.
- I, Robot, by Isaac Asimov
Some scientists doubt computers will match human intelligence anytime soon. IBM’s Watson Health artificial intelligence program was designed to help with disease diagnosis. It didn’t work very well.
“IBM has encountered a fundamental mismatch between the way machines learn and the way doctors work.”
- Strickland, 2019
Despite all the money poured into AI so far, Meta’s chief AI scientist Yann LeCun adds:
"We're not to the point where our intelligent machines have as much common sense as a cat."
In the unfolding story of OpenAI parallels that of Prometheus and the gift he gave humanity, there is a delicate balance between innovation and responsibility and the events underscore the need for cautious progress in AI development. Harnessed and directed responsibly, AI has the potential to help humanity overcome every issue that plagues us, from climate change to resource scarcity to war.
Pursued irresponsibly, AI carries the risk that we as a species could be replaced.
We really need to get this right.
Rate Relief
Markets are building in a rate cut. The Financial Forecast Center projects U.S. T-Bills could fall a full 2% from 5.60% in October to 3.65% by next June. Canada’s rates would likely fall in tandem.
This has the stock market very excited.
Financial Forecast Center
Perversely, this is because growth is slowing. The U.S. is expected to see growth decline from 2.4% in 2023 to 0.8% in 2024, and Canada will follow at about 0.7% (Conference Board).
Personal savings are tapped out and Buy-Now-Pay-Later purchases are surging, as are credit card balances.
We often see a rally in stocks from U.S. Thanksgiving through Christmas, and the interest rate cuts are pushing the rally further.
Interest rate declines suggest a slowdown or recession next year – no surprise. The indicators also suggest it will be shallow and won’t result in big job losses.
What To Do
If rates are going to decline due to a slowdown, then a lot of term deposits are going to come due at just the wrong time – when rates are down.
Consider 2-year terms when you renew this year, just to maintain a high rate for a little longer.
In the 1970s, rates went down for a few years, then jumped right back up because governments could not control their spending.
Much like what is happening today. Governments just can’t stop spending.
Check your chequing and savings accounts. If you aren’t making 4% or more, call us.
