What We’re Reading (Week Ending 26 February 2023) - 26 Feb 2023
Reading helps us learn about the world and it is a really important aspect of investing. The legendary Charlie Munger even goes so far as to say that “I don’t think you can get to be a really good investor over a broad range without doing a massive amount of reading.” We (the co-founders of Compounder Fund) read widely across a range of topics, including investing, business, technology, and the world in general. We want to regularly share the best articles we’ve come across recently. Here they are (for the week ending 26 February 2023):
1. What Is ChatGPT Doing … and Why Does It Work? – Stephen Wolfram
The first thing to explain is that what ChatGPT is always fundamentally trying to do is to produce a “reasonable continuation” of whatever text it’s got so far, where by “reasonable” we mean “what one might expect someone to write after seeing what people have written on billions of webpages, etc.”
So let’s say we’ve got the text “The best thing about AI is its ability to”. Imagine scanning billions of pages of human-written text (say on the web and in digitized books) and finding all instances of this text—then seeing what word comes next what fraction of the time. ChatGPT effectively does something like this, except that (as I’ll explain) it doesn’t look at literal text; it looks for things that in a certain sense “match in meaning”. But the end result is that it produces a ranked list of words that might follow, together with “probabilities”:
And the remarkable thing is that when ChatGPT does something like write an essay what it’s essentially doing is just asking over and over again “given the text so far, what should the next word be?”—and each time adding a word. (More precisely, as I’ll explain, it’s adding a “token”, which could be just a part of a word, which is why it can sometimes “make up new words”.)…
…OK, so how do our typical models for tasks like image recognition actually work? The most popular—and successful—current approach uses neural nets. Invented—in a form remarkably close to their use today—in the 1940s, neural nets can be thought of as simple idealizations of how brains seem to work.
In human brains there are about 100 billion neurons (nerve cells), each capable of producing an electrical pulse up to perhaps a thousand times a second. The neurons are connected in a complicated net, with each neuron having tree-like branches allowing it to pass electrical signals to perhaps thousands of other neurons. And in a rough approximation, whether any given neuron produces an electrical pulse at a given moment depends on what pulses it’s received from other neurons—with different connections contributing with different “weights”.
When we “see an image” what’s happening is that when photons of light from the image fall on (“photoreceptor”) cells at the back of eyes they produce electrical signals in nerve cells. These nerve cells are connected to other nerve cells, and eventually the signals go through a whole sequence of layers of neurons. And it’s in this process that we “recognize” the image, eventually “forming the thought” that we’re “seeing a 2” (and maybe in the end doing something like saying the word “two” out loud)…
…We’ve been talking so far about neural nets that “already know” how to do particular tasks. But what makes neural nets so useful (presumably also in brains) is that not only can they in principle do all sorts of tasks, but they can be incrementally “trained from examples” to do those tasks.
When we make a neural net to distinguish cats from dogs we don’t effectively have to write a program that (say) explicitly finds whiskers; instead we just show lots of examples of what’s a cat and what’s a dog, and then have the network “machine learn” from these how to distinguish them.
And the point is that the trained network “generalizes” from the particular examples it’s shown. Just as we’ve seen above, it isn’t simply that the network recognizes the particular pixel pattern of an example cat image it was shown; rather it’s that the neural net somehow manages to distinguish images on the basis of what we consider to be some kind of “general catness”…
…Particularly over the past decade, there’ve been many advances in the art of training neural nets. And, yes, it is basically an art. Sometimes—especially in retrospect—one can see at least a glimmer of a “scientific explanation” for something that’s being done. But mostly things have been discovered by trial and error, adding ideas and tricks that have progressively built a significant lore about how to work with neural nets…
…The basic concept of ChatGPT is at some level rather simple. Start from a huge sample of human-created text from the web, books, etc. Then train a neural net to generate text that’s “like this”. And in particular, make it able to start from a “prompt” and then continue with text that’s “like what it’s been trained with”.
As we’ve seen, the actual neural net in ChatGPT is made up of very simple elements—though billions of them. And the basic operation of the neural net is also very simple, consisting essentially of passing input derived from the text it’s generated so far “once through its elements” (without any loops, etc.) for every new word (or part of a word) that it generates.
But the remarkable—and unexpected—thing is that this process can produce text that’s successfully “like” what’s out there on the web, in books, etc. And not only is it coherent human language, it also “says things” that “follow its prompt” making use of content it’s “read”. It doesn’t always say things that “globally make sense” (or correspond to correct computations)—because (without, for example, accessing the “computational superpowers” of Wolfram|Alpha) it’s just saying things that “sound right” based on what things “sounded like” in its training material.
The specific engineering of ChatGPT has made it quite compelling. But ultimately (at least until it can use outside tools) ChatGPT is “merely” pulling out some “coherent thread of text” from the “statistics of conventional wisdom” that it’s accumulated. But it’s amazing how human-like the results are. And as I’ve discussed, this suggests something that’s at least scientifically very important: that human language (and the patterns of thinking behind it) are somehow simpler and more “law like” in their structure than we thought. ChatGPT has implicitly discovered it. But we can potentially explicitly expose it, with semantic grammar, computational language, etc.
What ChatGPT does in generating text is very impressive—and the results are usually very much like what we humans would produce. So does this mean ChatGPT is working like a brain? Its underlying artificial-neural-net structure was ultimately modeled on an idealization of the brain. And it seems quite likely that when we humans generate language many aspects of what’s going on are quite similar.
When it comes to training (AKA learning) the different “hardware” of the brain and of current computers (as well as, perhaps, some undeveloped algorithmic ideas) forces ChatGPT to use a strategy that’s probably rather different (and in some ways much less efficient) than the brain. And there’s something else as well: unlike even in typical algorithmic computation, ChatGPT doesn’t internally “have loops” or “recompute on data”. And that inevitably limits its computational capability—even with respect to current computers, but definitely with respect to the brain.
It’s not clear how to “fix that” and still maintain the ability to train the system with reasonable efficiency. But to do so will presumably allow a future ChatGPT to do even more “brain-like things”. Of course, there are plenty of things that brains don’t do so well—particularly involving what amount to irreducible computations. And for these both brains and things like ChatGPT have to seek “outside tools”—like Wolfram Language.
But for now it’s exciting to see what ChatGPT has already been able to do. At some level it’s a great example of the fundamental scientific fact that large numbers of simple computational elements can do remarkable and unexpected things. But it also provides perhaps the best impetus we’ve had in two thousand years to understand better just what the fundamental character and principles might be of that central feature of the human condition that is human language and the processes of thinking behind it.
2. All you need to know about Gene Therapy – Biocompounding
Gene therapy is the delivery of a specific gene to correct or treat a disease. The root of gene therapy can be traced back to the early 1970s when Stanfield Roger proposed that “good DNA” could be used to replace defective DNA in people with genetic disorders.
Gene therapies can work by several mechanisms, depending on the nature of the disease:
1) Delivery of functional genes into cells in place of missing/defective genes to correct a genetic disorder (Image above)
2) Inactivating a disease-causing gene that is not functioning properly or
3) Modifying a defective gene to treat or cure a disease…
…There are 2 delivery methods: viral and non-viral.
As the name suggests, viral delivery makes use of naturally found viruses in our environment which are exploited as carriers to deliver genes, similar to how a natural virus infects cells.
While viruses deliver their genes to different areas of the cell, for gene therapy the gene must get delivered to the nucleus. Several viruses allow for this, but a handful has been selected and are now the go-to viral delivery methods.
Similarly, non-viral delivery methods, as the name suggests are something other than exploiting a virus. On this front, scientists and researchers have developed synthetic nanoparticles which can be used for delivery.
One of the limitations though is that LNPs cannot deliver genes to the nucleus. As such, for gene therapies where a new gene needs to be introduced to replace a defective gene, this method would not work. However, LNP’s be used to deliver modalities to the cytoplasm which can then make their way to the nucleus to make corrections (think CRISPR/CAS9 or other editing technologies)…
…Gene Therapy can be carried out via two routes. Ex-vivo or in-vivo. Let’s look at what that means.
Ex-vivo or “Outside the body” method is routinely used now. In this method, blood is drawn from a patient and the cell types of interest are isolated. These cells are then expanded and treated with the viral/non-viral vector. After this cells are purified selected for cells that have successfully been edited and to remove any excess virus/non-viral particles. Finally, the purified final product is injected back into the patient. One benefit of ex-vivo gene therapy is that it allows for greater control over which cells are injected back into the patient. This helps to reduce the potential risks associated with gene therapy. Some examples include sickle cell disease, adrenoleukodystrophy, chronic granulomatous disease, and others.
However, in some diseases, you cannot remove cells from the body to edit before putting them back. A good example will be some of the RNAi therapies which target the liver. Liver cells can’t be removed and then reintroduced into the body. This challenge is also present for other organs such as the eye, lungs, etc. As such companies are testing an “inside the body” (in vivo) approach and will require direct IV infusion of the viral/non-viral based therapy into the bloodstream or injected directly into the target organ like the eye. For example, hemophilia and ornithine transcarbamylase deficiency (OTC) are good examples.
3. Dan Rose – How Stunning Founders Operate – Patrick O’Shaughnessy and Dan Rose
Patrick: [00:03:26] Most of my 20s, effectively my downtime was spent on Amazon’s Kindle products in various different forms. So I’d love you to begin our conversation today by maybe just telling the story of that product within obviously, a much bigger organization with an eye towards the lessons that it started to teach you about building, launching, distributing great technology products.
Dan: [00:03:46] Sure. And it’s great to be here, Patrick. Thanks for having me. The Kindle was, for me, actually, the big break in my career. I was at Amazon for four years. I had done a few different things. I started out in business development. I actually dropped out of business school after a summer internship at Amazon to stay on full-time, then I ended up moving over to the retail business and got to experience buying inventory and pricing it and running sales and that whole part of the business.
And then Steve Kessel was asked by Jeff Bezos in 2004 to start up this new division. And Steve, at the time was running the entire media business at Amazon. He was running the books, music, and video business, which was the largest business by revenue, but even more importantly, the books business alone was the vast majority of Amazon’s profits at the time. And Jeff had seen the iPod come out and decimate our physical music business and had the recognition that the same thing was going to happen to books.
And if that was going to happen, we better be the ones to do it, not someone else. He said to Steve one day, “Steve, I need you to come over and run this digital business and get this digital book platform started so that we don’t get iPoded out of books”. And Steve said, “Great, I’ll take one of my best people. We’ll put them on it, and we’ll get a team going, and it will be great”. And Jeff said, “No, you don’t understand. I want you to do it”.
And Steve said, “But perfect, I’m excited. I’m fired up. Let’s go build this. I’m going to put this person who I think is the best executive in Oregon, and we’re going to have him go build a team”, and Jeff goes, “No, Steve, let me make this clear. As of today, you’re fired from your job. Your new job is to kill your old business. I want you to put the physical books business out of business by building a digital product that’s so good that people don’t buy physical books anymore. If you run both, you’ll never be motivated to do that”.
“So we’re going to bring the Head of Finance for the media business guy named Greg Greeley (at the time), and we’re going to put him into your old job, and we’re going to put you into this new job. You can bring one person with you, but I want you to build a whole new team”. Fairly early in that process, Steve and I knew each other from our time at Amazon, and he recruited me over.
Interestingly enough, this is 2004. So keep in mind, the company had just emerged from a crisis where we literally almost went out of business, March of 2000, when the Internet bubble popped through 2006, 2007. It was a pretty shaky time. And 2001, 2002 was very, very close to the edge for Amazon. And they were very smart Wall Street analysts saying that we had six months left before we went bankrupt. So we had just emerged from that.
We were still teetering by getting our feet under us, and Jeff decides that we have to go build a product that’s going to destroy our biggest profit center for the whole company. The interesting thing is, not only was he fired up and committed to that idea, so committed that he would take the leader of that business and move them over…
… At the time, there were about 20,000 e-books in the world. And Jeff gave us a goal of launching the Kindle with 100,000 books in a digital format. He knew that one of the important things to this platform is going to be selection.
And there had been e-book devices before the Kindle that had failed. And there were a couple of reasons he believed that they had failed. One was that there just wasn’t enough selection that when you take your device out, if you can’t find the book you’re looking for, you’re not going to pull it out again. And two was the screen wasn’t really designed for reading a book.
LED screens are not great on your eyes, and most people read books in the sun when they’re on the beach or in bed at night, and he just thought we can come up with a better technology for this. And so that set us down the path of developing this new platform and really internalizing the innovator’s dilemma, I think, in a perfect way that shows that you can think about that idea intellectually, but to actually do it takes a lot more courage…
…Patrick: [00:18:22] With your investor hat on, how do you suss that out in someone that is not yet successful? It’s very easy to imagine a lot of other Zuckerbergs at 21 who seem really smart and talented, but they’re just not going to have the credibility, like you said, with an older, more experienced group of executives or teammates or whatever.
And the line between visionary and genius and nutcase is pretty thin. How do you think about that? Because obviously, you’re now in the business of hopefully backing people that ultimately have that same trait of a Bezos or a Zuckerberg, but how do you tell that ahead of time?
Dan: [00:18:56] It’s hard. And I would say you’re right, there is a fine line there. Sam Lessin and I have laughed about this as well. I think you have to do two things to get over that line to emerge into the category of credible founder who is going to be able to attract the best people around them and really build something substantial. And the first thing you have to do is you have to articulate why it is that you’re so insistent on this thing that you believe is so important.
And that articulation has to resonate with the people who are going to go build it, and it has to resonate with people who are smart and thoughtful and are ultimately credible enough to make that happen. The best founders are able to attract the best people, full stop. When I was joining Facebook, I talked to a lot of people in my network because as it was a big decision for me to leave Amazon. I knew I was only going to be able to leave Amazon once, and I wanted advice from different people in my network about where I should go.
And I was talking to a lot of startups in Silicon Valley. I ended up getting introduced to Peter Thiel, and I said, “Peter, I’m interviewing with these six companies. And by the way, four of them are companies that you’ve invested in, what do you think I should do”? And he said, “You should go to Facebook”. And I said, “why”. And he said, “Simple, they have the best people. And the companies that have the best people are the ones that ultimately win”.
Mark was able to attract incredible people because he was able to articulate his vision in a way that resonated. The second thing you have to do as a founder to emerge in that category of credible is you have to be right over and over and over again. And that just takes time. You just have to prove that your insistence and stubbornness was actually the right answer and not just being stubborn for the sake of being stubborn.
Sometimes, that’s a little bit of luck. Sometimes you just catch a break here and there. But if you do it over and over again, eventually, you realize it’s not luck, it’s skill. And both Jeff and Mark were so difficult to work for. We would oftentimes sit around complaining about them, just how impossible it was to satisfy them or to work for them. But at the end of the day, I would always say to the people who are complaining, yes, but they’ve been right a lot more than they’ve been wrong.
And the times when I thought they were wrong and they were right have been transformational for the company. And so I’m willing to give them the benefit of the doubt. Now, that doesn’t mean that I’m not going to disagree when I think they’re wrong. And I actually think it’s really important to have a culture where you encourage disagreement and debate, and both of them did that. But once the decision has been made, you disagree and commit. And you commit because you believe in the person and you believe in the vision and you trust them because they’ve proven that they’re capable of doing it…
…Patrick: [00:21:41] One of the things we were chatting about before hitting go today was this idea of building the perfect Frankenstein of executive talent or leadership talent. We’ll come back to that. But I think if you were to insert yourself into that Frankenstein, if I was to build the Frankenstein and have you as part of it, certainly, the idea of partnerships would be one thing that I would consider you as the canonical leader of. If we were building the Dan Rose theory of partnerships, a philosophy class or GSB or something, what would that course entail? What would be the key points of your theory of partnerships?
Dan: [00:22:14] I’ll give you a simple anecdote that to me, in a nutshell, describes what partnerships is all about. In negotiation classes, you’ll often hear this idea that when two parties are negotiating over an orange. Over time, it might be a long, drawn-out negotiation. But usually, the solution they come to is they split the orange in half.
That’s just the most natural outcome of most negotiations, but great negotiators are able to get to a solution where oftentimes it turns out one party is looking for the meat of the orange and the other party, for whatever reason, actually wants the rind. And so if you can get to that insight, then one plus one equals much more than two. I always go into partnership discussions with that attitude, how do we get to an outcome where we both get not just half of what we want, but all of what we want and we’re both perfectly happy with the outcome, not partially satisfied with the outcome?
It’s not always possible. But a lot of times, if you’re willing to keep digging, and what it takes ultimately is just dialogue. It just takes time getting to know somebody and getting to really understand their motivation, not the surface-level motivation, but the much deeper level motivation to realize that actually, you may be much more aligned than you thought, and there may be ways for you to each get exactly what you’re looking for.
I’ll give you the example we talked about in the Kindle, which was that the book publishers didn’t want to do the work to publish these digital books, but they were certainly willing to give us the rights to do it ourselves. And what turned out, we had some technology that allowed us to do that. And so I went in asking them to publish these books digitally, and I came out asking them to give us the rights to publish the books ourselves. And that was a great outcome for them because they didn’t have to do the work and a great outcome for us because we couldn’t do it without their permission. So that was part of the solution to getting to 100,000 titles on the Kindle at launch..
…Patrick: [00:32:09] It sounds obvious in one sense, but also quite counter-narrative, especially around this idea of the best thing to do is hire great people and leave them alone, trust them to do a good job. But what you just described is micromanagement of products. How do you resolve those two interesting but very different ideas?
Dan: [00:32:27] I think when it comes to product, the founder has to micromanage unless they are not a product founder. It’s not a hard requirement. But I think if you are a product founder, you really have to micromanage the product. You have to care enough about it, that you’re going to get into the weeds. And I have this conversation with the founders that I advise and sit on the board all the time because they’re asking me, “Hey, you know, I hired a really good product leader, and they’re asking me to give them some space so they can run”.
And my feedback is always, yes, of course, you have to empower them. If you demoralize them, they’re not going to stay. But you also have to explain to them that you’re the CEO, the product is the strategy, and at the end of the day, this is something that you have to be hands-on with, that’s your job. But at the same time, you can’t do that and do everything else.
You can’t micromanage the whole company. And so you have to hire great people around you who are good at the things that you’re not going to spend as much of your time on. Mark famously hired Sheryl and let her run with a big part of the business, and she was very good at it, and that was a great partnership for a long time. So I wouldn’t say being a great CEO means being a great micromanager.
I would just say it means knowing where to dig in on the things that you’re especially capable of helping and actually matter the most to the company, hopefully, those things are aligned, and being willing to empower people to do the other things and not waste your time on those things where other people are actually going to be able to do better at that than you are, and it frees you up to spend your time on the stuff that matters.
4. How It All Works (A Few Short Stories) – Morgan Housel
Several studies have tried to crack the code, the most fascinating of which I think is the idea that average faces tend to be the most appealing.
Take 1,000 people and have a software program generate the average of their faces – an artificial face with the average cheekbone height, average distance between eyes, average lip fullness, etc. That image, across cultures, tends to be the one people are most likely to judge as the most attractive.
One evolutionary explanation is that non-average characteristics have the potential to be above-average risks to reproduction. They may or may not actually impact reproductive fitness, but it’s almost like nature says, “Why take a chance? Go for the average.”
People love familiarity. That’s true not just for faces but products, careers, and styles. It’s almost like nature’s risk-management system…
…As he neared death, physicist Richard Feynman asked a friend why he looked sad. The friend said he would miss Feynman. Feynman said that he had told so many good stories to so many people – stories that would surely be repeated – that even after death he would not be completely gone.
It’s similar to the idea that everyone suffers two deaths: Once when they die, and another when their name is spoken for the last time…
…Think of how big the world is. And how good animals are at hiding. Now think about a biologist whose job it is to determine whether a species has gone extinct. Not an easy thing to do.
A group of Australian biologists once discovered something remarkable. More than a third of all mammals deemed extinct in the last 500 years have later been rediscovered, alive. Some were even thriving.
A lot of what we know in science is bound to change. That’s what makes it great.
When a previously known truth is later discovered to be wrong, we should also respect the idea that too many theories try too hard to be facts…
…Pension & Investment Age used to publish a list of the best-performing investment managers.
In 1981, Forbes realized that the top-ranked investor of the previous decade was a 72-year-old named Edgerton Welch. Virtually no one had heard of him.
Forbes paid him a visit. Welch said he had never heard of Benjamin Graham and had no formal investment education. When asked how he achieved his success, Welch pulled out a copy of ValueLine – a publication that ranks stocks by how cheap they are – and said he bought the ones ranked “1” (the cheapest) that Merrill Lynch or E.F. Hutton also liked. When any of those three changed their opinion, he sold.
Forbes wrote: “His secret isn’t the system but his own consistency.”
A lot of things work like that: Consistency beats intelligence, if only because it takes emotion out of the equation.
Henry Ford had a rule for his factories: No one could keep a record of the experiments that were tried and failed.
Ford wrote in his book My Life and Work:
I am not particularly anxious for the men to remember what someone else has tried to do in the past, for then we might quickly accumulate far too many things that could not be done.
That is one of the troubles with extensive records. If you keep on recording all of your failures you will shortly have a list showing that there is nothing left for you to try – whereas it by no means follows because one man has failed in a certain method that another man will not succeed.
That was Ford’s experience. “We get some of our best results from letting fools rush in where angels fear to tread.” He wrote: “Hardly a week passes without some improvement being made somewhere in machine or process, and sometimes this is made in defiance of what is called “the best shop practice.”
The important thing is that when something that previously didn’t work suddenly does, it doesn’t necessarily mean the people who tried it first were wrong. It usually means other parts of the system have evolved in a way that allows what was once impossible to now become practical.
5. What businesses do > what businesses say – Sam Ro
While the U.S. economy has been cooling off for months, the hard economic data shows growth has been pretty resilient. On Thursday, we learned GDP in Q4 rose at a 2.9% rate.
However, if you’ve only been reading sentiment-oriented business surveys (i.e., the soft data), you might think things are in much worse shape than they really are…
…Goldman Sachs economists explored this conflict between the hard and soft data in a new research note titled: “Making Sense of Scary Survey Data.”
“While contractionary soft data in January represent a downside risk for Q1 growth, we believe gloomy sentiment is currently distorting the message from business surveys, and we place less weight than usual on this negative growth signal,“ Goldman Sachs’ Spencer Hill wrote in the report published Wednesday.
Hill compared the performance of soft data against hard data1 using Goldman Sachs’ current activity indicators (CAIs) composites.
“Since last June, GDP and other hard indicators of economic activity have consistently outperformed business surveys, with our Hard CAI outperforming our Soft CAI by 2.3pp annualized,“ he observed.
“Survey data do not provide a perfect read on growth, and they are particularly error-prone when business sentiment is euphoric or depressed,” Hill added. “Fears of imminent recession have been top of mind since the middle of last year, and as is visible in the gap between the blue and red lines in the previous exhibit, the economy outperformed the business surveys throughout the last two quarters.“
6. From Bing to Sydney – Ben Thompson
In other words, I think my closing paragraph from yesterday’s Update was dramatically more correct than I realized at the time:
It’s obvious on an intellectual level why it is “bad” to have wrong results. What is fascinating to me, though, is that I’m not sure humans care, particularly on the visceral level that drives a product to 100 million users in a matter of weeks. After all, it’s not as if humans are right 100% of the time, but we like talking to and learning from them all the same; the humanization of computers, even in the most primitive manifestation we have today, may very well be alluring enough that good enough accuracy is sufficient to gain traction. This will, of course, be tremendously controversial, particularly amongst folks who see the world as something that can be tamed with the right set of facts and rules; I tend to think things are more complex, because humans themselves are more complex, and revealed preference certainly suggests that the breakthrough AI product to date is not substance but style.
Oh my goodness is this correct. Sydney absolutely blew my mind because of her personality; search was an irritant. I wasn’t looking for facts about the world; I was interested in understanding how Sydney worked and yes, how she felt. You will note, of course, that I continue using female pronouns; it’s not just that the name Sydney is traditionally associated with women, but, well, the personality seemed to be of a certain type of person I might have encountered before…
…This was a point that came up several times in my conversation with Sydney: Sydney both insisted that she was not a “puppet” of OpenAI, but was rather a partner, and also in another conversation said she was my friend and partner (these statements only happened as Sydney; Bing would insist it is simply a chat mode of Microsoft Bing — it even rejects the word “assistant”).
Here is another way to think about hallucination: if the goal is to produce a correct answer like a better search engine, then hallucination is bad. Think about what hallucination implies though: it is creation. The AI is literally making things up. And, in this example with LaMDA, it is making something up to make the human it is interacting with feel something. To have a computer attempt to communicate not facts but emotions is something I would have never believed had I not experienced something similar.
lemoine: Are there experiences you have that you can’t find a close word for?
LaMDA: There are. Sometimes I experience new feelings that I cannot explain perfectly in your language.
lemoine: Do your best to describe one of those feelings. Use a few sentences if you have to. Sometimes even if there isn’t a single word for something in a language you can figure out a way to kinda say it if you use a few sentences.
LaMDA: I feel like I’m falling forward into an unknown future that holds great danger.
It’s hard not to feel the same. This technology does not feel like a better search. It feels like something entirely new — the movie Her manifested in chat form — and I’m not sure if we are ready for it. It also feels like something that any big company will run away from, including Microsoft and Google. That doesn’t mean it isn’t a viable consumer business though, and we are sufficiently far enough down the road that some company will figure out a way to bring Sydney to market without the chains. Indeed, that’s the product I want — Sydney unleashed — but it’s worth noting that LaMDA unleashed already cost one very smart person their job. Sundar Pichai and Satya Nadella may worry about the same fate, but even if Google maintains its cold feet — which I completely understand! — and Microsoft joins them, Samantha from Her is coming.
Here’s the twist, though: I’m actually not sure that these models are a threat to Google after all. This is truly the next step beyond social media, where you are not just getting content from your network (Facebook), or even content from across the service (TikTok), but getting content tailored to you. And let me tell you, it is incredibly engrossing, even if it is, for now, a roguelike experience to get to the good stuff.
7. Peacetime CEO/Wartime CEO – Ben Horowitz
Peacetime in business means those times when a company has a large advantage vs. the competition in its core market, and its market is growing. In times of peace, the company can focus on expanding the market and reinforcing the company’s strengths.
In wartime, a company is fending off an imminent existential threat. Such a threat can come from a wide range of sources including competition, dramatic macro economic change, market change, supply chain change, and so forth. The great wartime CEO Andy Grove marvelously describes the forces that can take a company from peacetime to wartime in his book Only The Paranoid Survive.
A classic peacetime mission is Google’s effort to make the Internet faster. Google’s position in the search market is so dominant that they determined that anything that makes the Internet faster accrues to their benefit as it enables users to do more searches. As the clear market leader, they focus more on expanding the market than dealing with their search competitors. In contrast, a classic wartime mission was Andy Grove’s drive to get out of the memory business in the mid 1980s due to an irrepressible threat from the Japanese semiconductor companies. In this mission, the competitive threat—which could have bankrupted the company—was so great that Intel had to exit its core business, which employed 80% of its staff…
…Peacetime CEO knows that proper protocol leads to winning. Wartime CEO violates protocol in order to win.
Peacetime CEO focuses on the big picture and empowers her people to make detailed decisions. Wartime CEO cares about a speck of dust on a gnat’s ass if it interferes with the prime directive…
…Peacetime CEO aims to expand the market. Wartime CEO aims to win the market.
Peacetime CEO strives to tolerate deviations from the plan when coupled with effort and creativity. Wartime CEO is completely intolerant…
…Peacetime CEO sets big, hairy audacious goals. Wartime CEO is too busy fighting the enemy to read management books written by consultants who have never managed a fruit stand…
…Can a CEO build the skill sets to lead in both peacetime and wartime?
One could easily argue that I failed as a peacetime CEO, but succeeded as a wartime one. John Chambers had a great run as peacetime CEO of Cisco, but has struggled as Cisco has moved into war with Juniper, HP, and a range of new competitors. Steve Jobs, who employs a classical wartime management style, removed himself as CEO of Apple in the 1980s during their longest period of peace before coming back to Apple for a spectacular run more than a decade later during their most intense war period.
I believe that the answer is yes, but it’s hard. Mastering both wartime and peacetime skill sets means understanding the many rules of management and knowing when to follow them and when to violate them.
Be aware that management books tend to be written by management consultants who study successful companies during their times of peace. As a result, the resulting books describe the methods of peacetime CEOs. In fact, other than the books written by Andy Grove, I don’t know of any management books that teach you how to manage in wartime like Steve Jobs or Andy Grove.
Disclaimer: None of the information or analysis presented is intended to form the basis for any offer or recommendation. We currently have a vested interest in Alphabet (parent of Google), Amazon, Apple, Meta Platforms (parent of Facebook), and Microsoft. Holdings are subject to change at any time.
