Today, I am sharing the final part of my five-part summer letter to IMA clients. The letter is quite long, so I have split it into several parts to make it easier to read. In the first four parts, I have covered the economy, the Magnificent 7, past booms and busts, and EV. In this part, I will talk about AI, starting with the question:
Q: AI is transforming many industries. Citi released a report predicting that 54% of jobs in banking could be automated by AI. How can IMA invest to capitalize on AI disruption?
In these client letters, I am not trying to sell anything. They are written for IMA clients who are already aligned with what we do. I don't like to rewrite the letters (rewrite them into articles) because I don't think I would learn anything by doing that, so I'm leaving them as they are.
AI Revolution
Now, let's look at AI.
Let's state the obvious: discussions of AI quickly descend into science fiction realm, and my thoughts here are only lightly shaped by scientific fact.
When we look at technological advances over the centuries through the lens of productivity, the gains have been minimal and barely noticeable: our ability to do the same work might have improved by only a few basis points per year for millennia. People figured out fire, learned how to use levers, and invented the wheel.
Then, with the First and Second Industrial Revolutions, the pace of technological advances—steam engines, internal combustion engines, electricity, microprocessors, and the Internet—began to accelerate, and productivity went from improving by a few percent per century to improving by a few percent per year. The great thing about increased productivity is that it's the magic that makes our lives better. Technology allows us to use fewer resources to grow tomatoes or build a new house.
Coincidentally, while writing this article, I received an email from Edward Chancellor, who, in his article titled “The Trouble with Prosperity,” argues that productivity growth has slowed from 3% to 1% since the 1970s. Edward believes that the main causes are (1) prosperity making people less willing to work the richer they are, (2) governments playing a larger role in the economy (Western governments have been running deficits for nearly 50 years), and (3) a tenfold increase in regulation.
Hopefully AI can reverse this. Back to the future.
I've been battling here between two opposing frameworks.
On the one hand, technological advances over the past century have changed the nature of employment: Where once a third of the population worked on farms, today that proportion is tiny, with a single combine harvester replacing dozens of workers.
The story continues
While this transition has certainly been painful for many farmers, it has been a positive one for society as a whole. While the transition is happening, we can only see the jobs that will be lost, not the jobs that will be created, because those professions have not yet been invented. Think of the jobs that we take for granted today that didn't exist 50 or 100 years ago: laparoscopic surgeons, data scientists, chief listening officers. No joke, this was suggested by ChatGPT (who can probably see into the future).
On the other hand, as Larry Summers has pointed out, the biggest difference between AI and all other advances is that a wheel can't make another wheel, but AI can make AI. It's entirely possible that AI will exponentially increase growth and accelerate the rate of productivity growth. Again, I'm in science fiction territory, but it's not science fiction territory.
How AI will impact the job market
Most of the past technological innovations have hurt blue-collar jobs. The only exception I can think of is Xerox (NASDAQ:XRX) replacing a room with a female typing pool with a single copy machine. AI is color blind because it is coming to both blue-collar and white-collar jobs. I wanted to say that repetitive pattern recognition tasks will be swallowed up by AI, but ChatGPT has shown that AI can generate content. So creative jobs are not safe either. At the very least, productive outcomes will improve. In other words, AI tools will allow designers to do the work of two people or pick the numbers.
With robots and automation, plus AI, humans in warehouses will simply flip a switch. AI's attack on blue-collar jobs doesn't require much imagination. Amazon (NASDAQ:AMZN) and others are moving fast in that direction. Robots will start restocking supermarket shelves overnight. The Amazon warehouse workers who complain about having to walk 10 miles a day will start complaining that there are no new movies to watch on Netflix (NASDAQ:NFLX) because while they're out of work and flipping channels, robots are roaming and shipping.
The skill mix in the United States will change over time, and low-skilled workers will have a hard time finding jobs. (Plumbers, electricians, and other jobs that require manual skills are stable indefinitely.) Today, the United States imports many low-skilled workers through legal and illegal immigration. In ten years, these people will have a hard time finding jobs. (This is an economic view, not a political one.)
Uber and the Robotaxi Threat
Currently, there is only one company in our portfolio facing potential disruption from AI Uber (NYSE:UBER). The threat is a hypothetical scenario where millions of robotaxis suddenly appear on the roads. Uber's value comes from its fragmented, two-way network. If the supply of cars on the road becomes monopolized by a few large conglomerates (e.g. Tesla (NASDAQ:TSLA) or rental car companies), Uber's value proposition is significantly diminished.
But robotaxis are more science fiction than fact at this point: The technology isn't perfected and only works in select areas, and while Tesla's self-driving software has made great strides recently, it's still a long way from being able to confidently input a destination and then kill time on TikTok for the rest of the journey.
Self-driving technology relies heavily on sensors, which have limitations. For example, my Tesla's driver assistance feature stops working on rainy or snowy days when the cameras can't see the road clearly. Then there's the trust factor: many of us would be happy to pay a premium for the assurance that a human driver will get us safely from point A to point B, rather than taking the risk of getting into an empty self-driving car. There is also no legal framework for self-driving cars. If an accident occurs, who is to blame, this robot taxi or that robot taxi?
Tesla's robotaxi service is unlikely to succeed on its own. Tesla would struggle to maintain an abundant supply on its own, but this would make Uber's established network even more valuable. Consumers don't specifically want robotaxi; they want reliable transportation that arrives quickly when called (keyword).
As robotaxis slowly start to appear on the roads over the next few years, they will likely be a separate option on the Uber app (rather than operating independently), alongside UberX, Comfort, or Luxury options.
The social impact of AI
What will happen to our economy and society when AI takes over our offices and bots manage everything else? Let's say we introduce some kind of basic income system and people no longer need to work. How will that affect our society? Will people lose the meaning of life that is filled by work today?
But maybe I'm looking too far into the future, and in the meantime, AI is sure to bring about big changes.
I teach my kids that it's not about what AI does to us, but what we do to it. Our job is to embrace AI with an open mind and see it as a friend, not a threat. AI is going to happen whether we like it or not, so we should benefit from it.
From an investment perspective, there are plenty of companies that could benefit from AI, from robotics to automation to energy. The problem is, these companies are currently as fully valued as the rest of the market.
Vitaly Katzenelson is CEO of IMA, a value investing firm in Denver. He has written two books on investing, published by John Wiley & Sons and translated into eight languages. Soul in the Game: The Art of a Meaningful Life (Harriman House, 2022) is his first non-investing book. Forward your purchase receipt to [email protected] to receive an unpublished bonus chapter.
Please read our important disclosures here:
This article was originally published on GuruFocus.