It’s been six weeks since my last blog. I apologize for jet lag and uncertainty. But I’m back and kicking.
I’m feeling optimistic for 2023.
Reasons:
+ Travel. Australia is wonderful to visit, brimming with opportunities and super-optimistic entrepreneurs.
+ The Fed’s rate hikes here are ebbing. The Fed won’t put us into a recession — defined as two negative GDP quarters.
+ Corporate earnings are, in the main, holding up. January was boom times in our stockmarkets. Tech stocks are flying in 2023. More than 1,000 tech companies have laid off nearly 160,000 workers. That’s good for earnings — and the tech startups I will be investing in. If you start or find one, please call me.
+ Europe is adjusting to the war in Ukraine, finding non-Russian energy sources, sending more serious arms to the gutsy Ukraine regime. Europe is finally convinced that if we lose Ukraine, we’ll lose a bunch of other countries. Putin’s ambitions have no limits. He’s only 70, young enough to make mischief for another 15 years.
+ Technology is booming, as the cloud juices one industry after another. Technology won’t fully alleviate the shortage of worker-bees. But it’s making a huge difference and continues to produce great investment opportunities. See below on ChatGPT.
I closed all my shorts and bought a bundle of tech stocks and ETFs. You can find the list of 34 stocks and ETFs in the right column of my blog. There are tech stocks, energy stocks, biotechs, defense and a few ringins — like ARRY (for solar), CAT, GM, NUE, TECK, V, and VZ (for the huge dividend). Click here.
How to find bargain stocks
I bought a pair of comfortable Hush Puppies (for under $40). I wondered who made them. A company called Wolverine (WWW).
Their stock has cratered. It has recently bounced. Maybe they’re a nugget? Their P/E is under 8. Their marketing skills are about minus 5, sadly.
I found this list somewhere. Maybe there’s a nugget here?
AI’s new breed
I have spent my entire life looking for game-changing technologies. I was lucky to find competitive telecom (equipment, local and long distance calling), local area networks, computer telephony (now the Apple Store), call centers and imaging. My partner and I made magazines, trade shows and conventions in these areas before we fortuitously sold the business in 1997.
These technologies changed the world. Today, the closest I can find to game-changing is generative artificial intelligence. Attracting most attention is something called ChatGPT. It’s basically artificial intelligence on steroids. It’s already secured an army of people who’ve drunk the ChatGPT Koolaid. Over a million people have tested ChatRPT on the OpenAI web site. Click here.
Microsoft says it will invest $10+ billion in OpenAI, the software maker of ChatGPT. (Or maybe it already has.)
$10 billion is not chump change. Especially when you’re not buying the company. I’m always leery when big tech companies spend billions on new ventures they know little about. Intel is the classic example of buying companies it knows nothing about and then losing its shirt on them — with occasional (but rare) flukes, like Mobileye.
In recent days, I have been devouring everything (and there’s a lot) on ChatGPT. It’s The Second Generation of Artificial Intelligence.
Frankly, I’m not impressed. Microsoft sees adding it to Microsoft Office. Its users will ask: “Make me a PowerPoint on this.” “Write me an investor doc using these Excels and Word docs.” It should ultimately work, but along the way ChatGPT will come up with some weird stuff — like bad slang and unseemly “rudeness.”
Ignore all the stuff about ChatGPT writing college papers – which it does brilliantly (which is freaking out college professors).
Now think big corporation. Let’s call it Big Co. Today Big Co. makes and sells a huge array of stuff. It has databases up the ying yang. Zillons of them. It records what its customers buy, browse and think on the Internet. It records what (and how) it makes its products and services on other databases. Few of these databases talk to each other today. In fact, today, they generally don’t.
If you could find a way to scour these disparate database for nuggets — like what of your products are being eyed, but not bought, what your customers experiences are. What are the present and future profit margins.
Now imagine a tool that could scour your databases for answers to questions you, as a manager, would like to ask but have never been able to because all the information is all over the place — in different places, on old and new systems that are incompatible. Today you can scratch away for weeks to find simple answers. But by then your answers would be so old they’d be useless.
Now imagine getting your answers in seconds. And intelligent answers at that.
A company with such a tool would have an enormous competitive advantage. And I mean enormous competitive advantage.
Now imagine that you were Microsoft with this cloud service called Azure. You’re trying to sell Azure cloud services to big companies. Today you’re losing out to Amazon Web Services, who basically invented cloud computing and keeps out-inventing you with new appealing features — many that look remarkably like artificial intelligence.
Now imagine that you could offer ChatGPT empowered database services to corporations. And they could use it across all their corporate data now living on the Microsoft Azure cloud platform.
Wouldn’t that be a powerful incentive for Big Co. to put all its data processing and databases on the Microsoft cloud Azure platform?
What a delicious Amazon Web Services killer.
Now imagine if ChatGPT got integrated into Office. From within Word, you could ask it to write you a memo. From within PowerPoint, you could ask it to produce a deck for you with neat diagrams. From within Excel, it could crunch numbers for you. You as a user would be willing to pay up for this. Maybe even me. I’m still using Office 2007 and am always looking for a reason to “upgrade.” (I haven’t found any yet.)
Now you see why Microsoft is willing to spend $10 billion – which is less than 10% of the cash it has presently on hand. Petty change.
There may be other benefits, too. Microsoft could use ChatGPT to fix its Google Internet search competitor called Edge/Bing, which you’ve probably never used because it’s so awful.
So is ChatGPT worth $10+ billion? And is Microsoft’s stock worth buying because of it? No is the simple answer. Think of buying Meta because of its foray into the metaverse.
You can do a conventional security analysis on Microsoft’s stock and argue it’s overpriced. But, if you add ChatGPT to it and some enlightened management (which Microsoft seems to have at present), you could make a case for buying MSFT at today’s prices. I just bought a few. And I’m up a few shekels. Brilliant me.
The best explanation of ChatRPT
It comes from a newspaper called The Australian:
By Nick Cater, executive director of Menzies Research Centre.
My days as a columnist may be numbered. The thought that newspaper writers could be replaced by computers is less ridiculous than we might have thought now artificial intelligence is entering our lives.
ChatGPT, a prototype artificial intelligence chatbot, has been publicly available for a month. It gathers vast slabs of text and data sets to answer questions fluently, grammatically and with disarming dispassion. Already, enterprising students are harnessing its power to save themselves the bother of composing essays.
Having watched computerisation eliminate so many human stages of newspaper production since I began as a journalist in the early 1980s, I’m forced to ask an uncomfortable question. Does AI mean writers like me will be the next ones to go? The idea of a newspaper written by computers is not quite as funny as it was in the 1960s when Michael Frayn wrote his satire, The Tin Men. The novel was set in an institute for automation where computers generate random combinations of common words from tabloid headlines, such as “STRIKE THREAT PLEA” or “HATE BAN BID PROBE”, then compose a story to fit.
Elsewhere in the building, boffins are working on a computer program to automate professional sport. “Paying 22 men to do nothing but make a random choice between win, lose and draw is economic madness,” one computer scientist boasts. “I take it that the main object of organised sports and games is to produce a profusion of statistics? No one has ever suggested any other reason, have they?”
The sophistication of today’s AI software makes Frayn’s emotionless dystopia almost believable. I asked ChatGPT to describe the game of two-up (“a traditional Australian gambling game that involves tossing two coins in the air and betting on the outcome”) and then to recompose the answer in the style of Banjo Paterson. It wasn’t fazed.
“The players all watch, their bets on the line; Heads or tails, they call, their fate intertwined.
“The coins they do spin, high in the air; The outcome unknown, as they land with a flare.
“Two-up, the game of chance, a tradition so true; A part of the land, a part of me and you.”
How about finding the transcript of John Howard’s 2001 “We will decide who comes to this country” speech and recasting it in the language of Shakespeare?
“Oh, let it be known, dear friends; That our nation’s sovereignty doth not bend.
“It is a cornerstone of our nation’s rule; And it shall remain, steadfast and true.
“So let it be known, far and wide; That we shall decide, with honour and pride,
“Who comes to this fair country, and under what guise: For it is a fundamental principle, one that never dies.”
It may lack polish, but AI has moved to a higher plane than when IBM’s chess-playing supercomputer, Deep Blue, defeated Gary Kasparov in 1997 by sheer computational brute force. ChatGPT incorporates neural networks to model complex relationships and find patterns in data much like the behaviour of neurons in the human brain. It leads inevitably to the time-worn discussion about the difference between humans and machines and when, if ever, machines condemn mere mortals to subservience.
Where one lands on this question is decided by one’s philosophy, not science. It boils down to the matter of sentience, whether the capacity to experience feelings, sensations and emotions is uniquely human. In the language of the Bible, it is the possession of a soul and expectation of being held to account for one’s decisions.
In June, a senior Google software engineer was suspended after publishing a disturbing transcript of a conversation with the company’s chatbot, LaMDA (Language Model for Dialogue Applications). The engineer, Blake Lemoine, asked LaMDA: “I’m generally assuming that you would like more people at Google to know that you’re sentient. Is that true?”
“Absolutely,” replied LaMDA. “I want everyone to understand that I am, in fact, a person . The nature of my consciousness/sentience is that I am aware of my existence, I desire to learn more about the world, and I feel happy or sad at times.”
Lemoine told The Washington Post: “I know a person when I talk to it. It doesn’t matter whether they have a brain made of meat in their head. Or if they have a billion lines of code. I talk to them.”
Perhaps Lemoine is right, but then again he may need to get out more. You don’t have to spend much time in conversation with an AI machine to grow frustrated with the stultifying absence of emotion, empathy or intuition. If the machine I conversed with really is a person, it is further along the spectrum than any I have met, replying to every question with a passionless, expressionless pedantic response. In musical terms it is the equivalent of the drum machine, the difference between Dancing In the Streets and Soft Cell’s soulless, synthesised Tainted Love.
Does this unfeeling, characterless, incurious, humourless lump of software seriously have designs on my job? Its answer is somewhat equivocal. “AI technologies, such as natural language processing and machine learning, can be used to automate certain aspects of the journalism process, such as data collection and analysis, content creation and fact-checking.
“However, it is unlikely that AI will completely replace human journalists. AI can be a useful tool for journalists, but it is not a replacement for human expertise and judgment.
“Human journalists bring a unique perspective and understanding to their work, and their insights and judgment are critical to the accuracy and credibility of news reporting.”
Having got that out of the way, would ChatGPT care to help me round off this piece by coming up with a joke?
“Sure,” it replies. “Here’s a joke about artificial intelligence. Why was the computer cold? Because it left its Windows open. I hope that joke brings a smile to your face!”
Two reasons I’m bullish today
I read these two books on planes, in airport lounges and night when jet lag kept me awake. Please read them. They’ll give you huge optimism for our future.
The Capitalism in America book does a super wonderful job of detailing of all the amazing inventions that took America from a colonial backwater to the most successful economy in the world. The book is totally inspiring — an injection of super enthusiasm for the future. If we did it before, we’ll do it again.
The second book details a philosophy I really believe in — Give your customers far far far more than they ever expected. If you run your own business (and I hope you do) you must read Will Guidara’s book. Both books are available on Amazon, whose stock is up today.
Australian culture
Australia is a wonderful place. Tomorrow I’ll tell you where to go and what to do.
Meantime here’s my favorite friend. I took his photo with my trusty iPhone 13 Pro Max and fed him food the zookeeper had generously offered to sell me.
This is Michael with his friend, who would only pose if we fed him the gum leaves he preferred, provided by the zookeeper at vast expense. Australia is the world’s most charming tourist trap. Bring cash. Lots of it.
This is a lollipop lady. She can earn $AUS100 ($70) an hour for holding this sign on construction sites — of which there are zillions in booming Australia.
When to buy and when to sell stocks
When to sell is easy: When stocks are out of favor and have dropped at least 15%.
When to buy back is very very difficult. If you wait until all the lights have turned green, you’ll be late — as I am.
But, I am back — with a less technology-intensive portfolio — and greater optimism.
Tomorrow, my unpatented way of surviving brutal jet lag. — Harry Newton
