We’ve constantly been sharing a list of our recent reads in our weekly emails for The Good Investors.

Do subscribe for our weekly updates through the orange box in the blog (it’s on the side if you’re using a computer, and all the way at the bottom if you’re using mobile) – it’s free!

But since our readership-audience for The Good Investors is wider than our subscriber base, we think sharing the reading list regularly on the blog itself can benefit even more people. The articles we share touch on a wide range of topics, including investing, business, and the world in general. 

Here are the articles for the week ending 12 January 2025:

1. The art of outlasting: What we can learn from timeproof Japanese businesses – Eric Markowitz

Japan is home to an extraordinary number of shinise, or long-established businesses. A 2008 study found that Japan had over 21,000 companies older than 100 years, including more than 3,000 that had crossed the 200-year mark. These firms are not just historical artifacts — they are vibrant examples of how to endure and thrive in a rapidly changing world. Their strategies — balancing tradition with adaptability, patience with practicality — are a masterclass in long-term thinking that today’s entrepreneurs and executives would be wise to study…

…What ties these stories together is an approach to business that’s almost rebellious in its patience. While the modern world glorifies disruption and speed, Japan’s ancient companies remind us that longevity is often about playing the long game. It’s about building something so solid, so aligned with its environment, that it can weather any storm. But let’s not romanticize this too much. Strip away the poetry of water metaphors and ancient traditions, and you’ll find ruthless pragmatism at the core of these businesses’ survival.

When Japan’s post-war construction boom faded, Kongo Gumi didn’t just stick to temples — they pivoted hard into office buildings and apartments while maintaining their temple maintenance business as a hedge. During the lean years of the 1990s recession, Hōshi Ryokan cut costs to the bone while refusing to lay off staff, with family members taking deep pay cuts to keep their centuries-old workforce intact. Okaya transformed from selling samurai swords to becoming a global steel trader, making calculated bets on new technologies and markets while keeping their supply chain relationships rock solid.

These companies didn’t just drift through history — they clawed their way through wars, depressions, and cultural upheavals, making brutal choices about what to preserve and what to sacrifice. Their longevity wasn’t achieved through Zen-like detachment, but through gritted teeth and white-knuckled adaptability.

2. Notes on China – Dwarkesh Patel

I got quite mixed messages about the state of public opinion in China. This is to be expected in a society where you can’t establish common knowledge. One person told me that the new generation is quite nationalist, unlike the older reform generation which personally experienced the catastrophes of Mao and the tangible benefits of liberalization. He made the rather insightful point that this tilt in Chinese public opinion increasingly gives lie to the American talking point, “We’re against the CCP, not the Chinese people.” In fact, he went on to say that the current regime is way more liberal than what would result from an election in China.

Another person told me that these Chinese nationalists were only a vocal minority, similar to the wokes in America circa 2020. While they make up only about 10% of the population, they aggressively shout down others on Weibo (China’s Twitter equivalent). Most people find them annoying but feel uncomfortable confronting them directly. This matches what a student who graduated from a top university there told me – the vast majority of his classmates are simply apolitical. And in our own interactions with locals, we saw little evidence of widespread nationalism. In fact, when my Chinese-speaking trip mate (who could actually speak Chinese) would mention he was from the UK to taxi drivers, they would often respond enthusiastically: “Oh wonderful, we love the UK!”…

…We chatted up quite a lot of young people on night life streets. I was struck by how many young people expressed feeling stressed or overwhelmed. We met a musician in Chengdu who was writing songs about youth anxiety. We chatted up some modeling school students – even they complained about the intense pressure they felt. We met a guy who had studied in Australia but returned to China during COVID. He explained that many of his friends with prestigious degrees are moving away from Shanghai and Beijing – Yes, the pay there can be twice as high as in second or third tier cities. But the competitiveness is insane. And in order to actually land the high skilled positions, they have to work truly insane hours (9-9-6 is not a myth). He said that many of his friends were opting for these less ambitious lower-paying careers in smaller cities, where the rent is lower and the pressure is manageable…

…I’m still puzzled by how China can have both a demographic collapse and massive youth unemployment. You’d think with fewer young people being born, the ones who are around would be in high demand. One explanation I heard while there is that there are plenty of menial jobs available, but today’s educated youth – who’ve gone through high school and college – just won’t take the low-skilled positions their parents and grandparents did. Meanwhile, there’s a real shortage of the high-skilled jobs that would actually match their education and aspirations. It’s a mismatch between the jobs available and the jobs young people feel qualified for and willing to do…

…The biggest surprise from talking to Chinese VCs people at AI labs was how capital constrained they felt. Moonshot AI, one of China’s leading AI labs, raised $1 billion at a $3 billion valuation. Meanwhile, just xAI’s new cluster alone will cost $3-4 billion.

The tech ecosystem feels quite shell shocked from the 2021 crackdown. One VC half-jokingly asked if I could help him get his money out of China. If you keep your money in China, you’re basically stuck choosing between terrible options. You can either accept a measly 2% yield from state banks, or throw it into China’s perpetually struggling stock market. This helps explain why valuations for Chinese companies are chronically low – the exit opportunities just suck. Even if you build (or invest in) something great, there’s no guarantee the company will be able to raise the next round. And even if you do raise again and succeed, the government might randomly cancel your IPO. And even if you somehow make it to the public markets, Chinese equities have been performing terribly anyways. It’s a good reminder of how easy it is to completely wreck an innovation ecosystem that depends on risk-taking investors.

3. Is AI progress slowing down? – Arvind Narayanan and Sayash Kapoor

To be clear, there is no reason to doubt the reports saying that many AI labs have conducted larger training runs and yet not released the resulting models. But it is less clear what to conclude from it. Some possible reasons why bigger models haven’t been released include:

• Technical difficulties, such as convergence failures or complications in achieving fault tolerance in multi-datacenter training runs.
• The model was not much better than GPT-4 class models, and so would be too underwhelming to release.
• The model was not much better than GPT-4 class models, and so the developer has been spending a long time trying to eke out better performance through fine tuning.

To summarize, it’s possible that model scaling has indeed reached its limit, but it’s also possible that these hiccups are temporary and eventually one of the companies will find ways to overcome them, such as by fixing any technical difficulties and/or finding new data sources…

…Industry leaders don’t have a good track record of predicting AI developments. A good example is the overoptimism about self-driving cars for most of the last decade. (Autonomous driving is finally real, though Level 5 — full automation — doesn’t exist yet.) As an aside, in order to better understand the track record of insider predictions, it would be interesting to conduct a systematic analysis of all predictions about AI made in the last 10 years by prominent industry insiders.

There are some reasons why we might want to give more weight to insiders’ claims, but also important reasons to give less weight to them. Let’s analyze these one by one. It is true that industry insiders have proprietary information (such as the performance of as-yet-unreleased models) that might make their claims about the future more accurate. But given how many AI companies are close to the state of the art, including some that openly release model weights and share scientific insights, datasets, and other artifacts, we’re talking about an advantage of at most a few months, which is minor in the context of, say, 3-year forecasts.

Besides, we tend to overestimate how much additional information companies have on the inside — whether in terms of capability or (especially) in terms of safety. Insiders warned for a long time that “if only you know what we know…” but when whistleblowers finally came forward, it turns out that they were mostly relying on the same kind of speculation that everyone else does.

Another potential reason to give more weight to insiders is their technical expertise. We don’t think this is a strong reason: there is just as much AI expertise in academia as in industry. More importantly, deep technical expertise isn’t that important to support the kind of crude trend extrapolation that goes into AI forecasts. Nor is technical expertise enough — business and social factors play at least as big a role in determining the course of AI. In the case of self-driving cars, one such factor is the extent to which societies tolerate public roads being used for experimentation. In the case of large AI models, we’ve argued before that the most important factor is whether scaling will make business sense, not whether it is technically feasible…

…As an example, Sutskever had an incentive to talk up scaling when he was at OpenAI and the company needed to raise money. But now that he heads the startup Safe Superintelligence, he needs to convince investors that it can compete with OpenAI, Anthropic, Google, and others, despite having access to much less capital. Perhaps that is why he is now talking about running out of data for pre-training, as if it were some epiphany and not an endlessly repeated point.

To reiterate, we don’t know if model scaling has ended or not. But the industry’s sudden about-face has been so brazen that it should leave no doubt that insiders don’t have any kind of crystal ball and are making similar guesses as everyone else, and are further biased by being in a bubble and readily consuming the hype they sell to the world…

…Inference scaling is useful for problems that have clear correct answers, such as coding or mathematical problem solving. In such tasks, at least one of two related things tend to be true. First, symbolic reasoning can improve accuracy. This is something LLMs are bad at due to their statistical nature, but can overcome by using output tokens for reasoning, much like a person using pen and paper to work through a math problem. Second, it is easier to verify correct solutions than to generate them (sometimes aided by external verifiers, such as unit tests for coding or proof checkers for mathematical theorem proving).

In contrast, for tasks such as writing or language translation, it is hard to see how inference scaling can make a big difference, especially if the limitations are due to the training data. For example, if a model works poorly in translating to a low-resource language because it isn’t aware of idiomatic phrases in that language, the model can’t reason its way out of this.

The early evidence we have so far, while spotty, is consistent with this intuition. Focusing on OpenAI o1, it improves compared to state-of-the-art language models such as GPT-4o on coding, math, cybersecurity, planning in toy worlds, and various exams. Improvements in exam performance seem to strongly correlate with the importance of reasoning for answering questions, as opposed to knowledge or creativity: big improvements for math, physics and LSATs, smaller improvements for subjects like biology and econometrics, and negligible improvement for English.

Tasks where o1 doesn’t seem to lead to an improvement include writing, certain cybersecurity tasks (which we explain below), avoiding toxicity, and an interesting set of tasks at which thinking is known to make humans worse…

…We think there are two reasons why agents don’t seem to benefit from reasoning models. Such models require different prompting styles than regular models, and current agentic systems are optimized for prompting regular models. Second, as far as we know, reasoning models so far have not been trained using reinforcement learning in a setting where they receive feedback from the environment — be it code execution, shell interaction, or web search. In other words, their tool use ability is no better than the underlying model before learning to reason…

…The furious debate about whether there is a capability slowdown is ironic, because the link between capability increases and the real-world usefulness of AI is extremely weak. The development of AI-based applications lags far behind the increase of AI capabilities, so even existing AI capabilities remain greatly underutilized. One reason is the capability-reliability gap — even when a certain capability exists, it may not work reliably enough that you can take the human out of the loop and actually automate the task (imagine a food delivery app that only works 80% of the time). And the methods for improving reliability are often application-dependent and distinct from methods for improving capability. That said, reasoning models also seem to exhibit reliability improvements, which is exciting.

Here are a couple of analogies that help illustrate why it might take a decade or more to build products that fully take advantage of even current AI capabilities. The technology behind the internet and the web mostly solidified in the mid-90s. But it took 1-2 more decades to realize the potential of web apps. Or consider this thought-provoking essay that argues that we need to build GUIs for large language models, which will allow interacting with them with far higher bandwidth than through text. From this perspective, the current state of AI-based products is analogous to PCs before the GUI.

4. Waymo still doing better than humans at preventing injuries and property damage – Andrew J. Hawkins

The study is the product of the collaboration between Waymo and insurer Swiss Re, which analyzed liability claims related to collisions from 25.3 million fully autonomous miles driven by Waymo in four cities: Phoenix, San Francisco, Los Angeles, and Austin. They then compared those miles to human driver baselines, which are based on Swiss Re’s data from over 500,000 claims and over 200 billion miles traveled.

They found that the performance of Waymo’s vehicles was safer than that of humans, with an 88 percent reduction in property damage claims and a 92 percent reduction in bodily injury claims. Across 25.3 million miles, Waymo was involved in nine property damage claims and two bodily injury claims. The average human driving a similar distance would be expected to have 78 property damage and 26 bodily injury claims, the company says.

Waymo’s vehicles also performed better when compared to new vehicles equipped with all the latest safety tech, including automatic emergency braking, lane-keep assist, and blind spot detection. When compared to this group, Waymo’s autonomous driving system showed an 86 percent reduction in property damage claims and a 90 percent reduction in bodily injury claims.

5. SITALWeek #454 – Brad Slingerlend

I think we are approaching the point where we can start to estimate the value of AI for developers and the companies/consumers who are going to buy the next wave of innovative applications. I think the salient question for AI (and, frankly, humanity!) is: How much AI reasoning can you get for a human-equivalent salary? In other words, for a certain salary, how much compute power will it take to match or outperform a human (assuming the AI can collaborate with other humans/AIs using the same methods and tools a human would)…

… LLMs are shifting from a pure token-in/token-out model to a test-time scaling model, which may offer us better inroads for estimating costs. Essentially, they are thinking harder before spitting out a reply; thus, rather than just predicting the next words in a response using a probability model (see You Auto-Complete Me), they are doing some deep thinking to arrive at more accurate, useful answers. This is a major leap in capability that comes with a major leap in cost. OpenAI raised prices for their o1 model to $200/mo (Pro subscription) from $20 (Plus subscription). For developers, use of o1’s advanced reasoning API comes at 3-4x the cost of their “general purpose” GPT-4o. If o1 were priced at a typical Western office worker wage of $40/hr, the reasoning of the model would equate to around 5 hours of work per month. We also don’t know if the $200/mo price point is profitable for OpenAI or if they are just relying on Microsoft to further subsidize their business model (which brings us back to the principal-agent problem I started this section off with). So, all of my hand waving here seems to imply you can get a decent amount of human-equivalent reasoning for an amount of money in the realm of human labor cost. If true, after a few more years of advancements in semiconductors and AI models, we should have markedly affordable “human reasoning as a service”, an explosion in demand, and a wide range of outcomes for how much human supervision of AI will be required (it may be that human jobs stay relatively flat, but each human is 2x productive, then 4x, etc.).

Following this logic, at current AI reasoning costs, companies would need to lay off one human for every AI human equivalent they hire and would probably lose more skill/knowledge than they gain. In other words, based on my attempts to guess the cost of replacing human reasoning, today’s AI offerings aren’t likely compelling enough. In a couple years, however, maybe you will be able to lay off one human and hire a handful of AIs, which, by collaborating with each other and humans, may yield superior results. Even today, extremely high-value tasks, such as in-depth research or stock market predictions, may be able to take advantage of the high-cost test-time scaling AI models. And, if any of this math is in the realm of reason, you can easily see that AI may not require such high-value-add applications to be cost effective in the near to medium future. The proof will come within the next couple of years as today’s entrepreneurs develop the next generation of apps leveraging LLMs and overtaking human capabilities: If these apps are at price points that outcompete human employees, a significant wave of change could come much faster to society. 

---

Disclaimer: The Good Investors is the personal investing blog of two simple guys who are passionate about educating Singaporeans about stock market investing. By using this Site, you specifically agree that none of the information provided constitutes financial, investment, or other professional advice. It is only intended to provide education. Speak with a professional before making important decisions about your money, your professional life, or even your personal life. We currently have a vested interest in Alphabet (parent of Google and Waymo) and Microsoft. Holdings are subject to change at any time.

We’ve constantly been sharing a list of our recent reads in our weekly emails for The Good Investors.

Do subscribe for our weekly updates through the orange box in the blog (it’s on the side if you’re using a computer, and all the way at the bottom if you’re using mobile) – it’s free!

But since our readership-audience for The Good Investors is wider than our subscriber base, we think sharing the reading list regularly on the blog itself can benefit even more people. The articles we share touch on a wide range of topics, including investing, business, and the world in general. 

Here are the articles for the week ending 29 December 2024:

1. Quantum Computers Cross Critical Error Threshold – Ben Brubaker

In the 1990s, researchers worked out the theoretical foundations for a way to overcome these errors, called quantum error correction. The key idea was to coax a cluster of physical qubits to work together as a single high-quality “logical qubit.” The computer would then use many such logical qubits to perform calculations. They’d make that perfect machine by transmuting many faulty components into fewer reliable ones…

…This computational alchemy has its limits. If the physical qubits are too failure-prone, error correction is counterproductive — adding more physical qubits will make the logical qubits worse, not better. But if the error rate goes below a specific threshold, the balance tips: The more physical qubits you add, the more resilient each logical qubit becomes.

Now, in a paper(opens a new tab) published today in Nature, Newman and his colleagues at Google Quantum AI have finally crossed the threshold. They transformed a group of physical qubits into a single logical qubit, then showed that as they added more physical qubits to the group, the logical qubit’s error rate dropped sharply…

…At first, many researchers thought quantum error correction would be impossible. They were proved wrong in the mid-1990s, when researchers devised simple examples of quantum error-correcting codes. But that only changed the prognosis from hopeless to daunting.

When researchers worked out the details, they realized they’d have to get the error rate for every operation on physical qubits below 0.01% — only one in 10,000 could go wrong. And that would just get them to the threshold. They would actually need to go well beyond that — otherwise, the logical qubits’ error rates would decrease excruciatingly slowly as more physical qubits were added, and error correction would never work in practice…

…That variation, called the surface code, is based on two overlapping grids of physical qubits. The ones in the first grid are “data” qubits. These collectively encode a single logical qubit. Those in the second are “measurement” qubits. These allow researchers to snoop for errors indirectly, without disturbing the computation.

This is a lot of qubits. But the surface code has other advantages. Its error-checking scheme is much simpler than those of competing quantum codes. It also only involves interactions between neighboring qubits — the feature that Preskill found so appealing.

In the years that followed, Kitaev, Preskill and a handful of colleagues fleshed out the details(opens a new tab) of the surface code. In 2006, two researchers showed(opens a new tab) that an optimized version of the code had an error threshold around 1%, 100 times higher than the thresholds of earlier quantum codes. These error rates were still out of reach for the rudimentary qubits of the mid-2000s, but they no longer seemed so unattainable…

…Fowler, Martinis and two other researchers wrote a 50-page paper(opens a new tab) that outlined a practical implementation of the surface code. They estimated that with enough clever engineering, they’d eventually be able to reduce the error rates of their physical qubits to 0.1%, far below the surface-code threshold. Then in principle they could scale up the size of the grid to reduce the error rate of the logical qubits to an arbitrarily low level. It was a blueprint for a full-scale quantum computer…

…When you put the theory of quantum computing into practice, the first step is perhaps the most consequential: What hardware do you use? Many different physical systems can serve as qubits, and each has different strengths and weaknesses. Martinis and his colleagues specialized in so-called superconducting qubits, which are tiny electrical circuits made of superconducting metal on silicon chips. A single chip can host many qubits arranged in a grid — precisely the layout the surface code demands.

The Google Quantum AI team spent years improving their qubit design and fabrication procedures, scaling up from a handful of qubits to dozens, and honing their ability to manipulate many qubits at once. In 2021, they were finally ready to try error correction with the surface code for the first time. They knew they could build individual physical qubits with error rates below the surface-code threshold. But they had to see if those qubits could work together to make a logical qubit that was better than the sum of its parts. Specifically, they needed to show that as they scaled up the code — by using a larger patch of the physical-qubit grid to encode the logical qubit — the error rate would get lower.

They started with the smallest possible surface code, called a “distance-3” code, which uses a 3-by-3 grid of physical qubits to encode one logical qubit (plus another eight qubits for measurement, for a total of 17). Then they took one step up, to a distance-5 surface code, which has 49 total qubits. (Only odd code distances are useful.)

In a 2023 paper(opens a new tab), the team reported that the error rate of the distance-5 code was ever so slightly lower than that of the distance-3 code. It was an encouraging result, but inconclusive — they couldn’t declare victory just yet…

…At the beginning of 2024, they had a brand-new 72-qubit chip, code-named Willow, to test out. They spent a few weeks setting up all the equipment needed to measure and manipulate qubits…

…Then a graph popped up on the screen. The error rate for the distance-5 code wasn’t marginally lower than that of the distance-3 code. It was down by 40%. Over the following months, the team improved that number to 50%: One step up in code distance cut the logical qubit’s error rate in half…

…The team also wanted to see what would happen when they continued to scale up. But a distance-7 code would need 97 total qubits, more than the total number on their chip. In August, a new batch of 105-qubit Willow chips came out…

…When the group returned the following morning, they saw that going from a distance-5 to a distance-7 code had once again cut the logical qubit’s error rate in half. This kind of exponential scaling — where the error rate drops by the same factor with each step up in code distance — is precisely what the theory predicts. It was an unambiguous sign that they’d reduced the physical qubits’ error rates well below the surface-code threshold…

…At the same time, researchers recognize that they still have a long way to go. The Google Quantum AI team only demonstrated error correction using a single logical qubit. Adding interactions between multiple logical qubits will introduce new experimental challenges.

Then there’s the matter of scaling up. To get the error rates low enough to do useful quantum computations, researchers will need to further improve their physical qubits. They’ll also need to make logical qubits out of something much larger than a distance-7 code. Finally, they’ll need to combine thousands of these logical qubits — more than a million physical qubits.

2. History: Kodak & Fujifilm – Find Value

Ultimately, Kodak couldn’t adapt to the changing world and filed for bankruptcy in 2012.

In the game for over 100 years, Kodak survived two World Wars and the Great Depression and helped humans photograph the moon and Mars. Like Coca-Cola and McDonald’s, it used to be one of the most recognized brands in the world…

…Faced with a sharp decline in sales from its cash cow product, Fujifilm acted swiftly and changed its business through innovation and external growth. Under Shigetaka Komori (President in 2000), Fujifilm quickly carried out massive reforms. In 2004, Komori came up with a six-year plan called VISION75.

The management restructured its film business by downscaling the production lines and closing redundant facilities. In the meantime, the R&D departments moved to a newly built facility to unify the research efforts and promote better communication and innovation culture among engineers.

Realizing that the digital camera business would not replace the lucrative film due to the low margins, Fujifilm performed a massive diversification based on capabilities and innovation.

Even before launching the VISION75 plan, Komori had taken stock of their technologies and compared them with the demand of the international market. After which the R&D team came up with a chart listing the all existing in-house technologies that could match future markets.

For instance, Fujifilm was able to predict the boom of LCD screens and invested heavily in this market. Leveraging on photo film technology, they created FUJITAC, a variety of high-performance films essential for making LCD panels for TV, computers, and smartphones. Today, FUJITAC owns 70% of the market for protective LCD polarizer films.

Fujifilm also targeted unexpected markets like cosmetics. The rationale behind cosmetics comes from 70 years of experience in gelatin, the chief ingredient of photo film which is derived from collagen. Human skin is 70% collagen. Fujifilm also possessed deep knowledge in oxidation, a process connected both to the aging of human skin and to the fading of photos over time.

When promising technologies didn’t exist internally, Fujifilm proceeded by mergers and acquisitions. Based on technological synergies, it acquired Toyoma Chemical in 2008 to enter the drug business. Delving further into the healthcare segment, Fujifilm also brought a radio-pharmaceutical company now called Fujifilm RI Pharma. It also reinforced its position in existing joint ventures such as Fuji-Xerox which became a consolidated subsidiary in 2001 after Fujifilm purchased an additional 25% share in this partnership.

Fast forward 9 years after the peak of film sales, in 2010, Fujifilm was a new company. In 2000, 60% of sales and 70% of profits came from the film ecosystem, compare this to 2010 where the “Imaging segment” accounted for less than 16% of sales. Fujifilm managed to emerge victorious through a restructuring and diversification strategy…

…Unlike Fujifilm which recognized early on that photography was a doomed business and tackled new markets with a completely different portfolio, Kodak made multiple wrong moves and persisted in the decaying film industry.

It was not that Kodak didn’t want to change, it tried hard, but it did it wrong. Kodak’s management didn’t fully recognize that the rise of digital imaging would have dire consequences for the future of photo printing. It tried to replicate the film print business model in the digital world. In 2004, Facebook was launched, and people are just not going to print pictures anymore.

Interestingly, Kodak understood the impact of digitalization and predicted that pictures would be shared online. They acquired a photo-sharing website called Ofoto in 2001. Unfortunately, the company used Ofoto to make people print digital pictures. They failed in realizing that online photo sharing was the new business, not just a way to expand printing sales…

…While Fujifilm invested heavily in the pharmaceutical and healthcare sector to reduce its exposure to the challenging photo industry, Kodak sold its highly profitable Healthcare Imaging branch in 2007 to put more resources into its losing consumer camera division.

3. One Bed, Two Dreams: Building Silicon Valley Bank in China with Ken Wilcox (Transcript here) – Bernard Leong and Ken Wilcox

Wilcox: In the US, banks sometimes fail. When I started my career 40 years ago in banking, we had 18,000 banks. Today we have about 5,000. What happened to all of them? Where did 13,000 banks go? Some of them got acquired, but many of them failed. When a bank makes too many bad loans, the Federal Reserve causes it to fail and it disappears. In China, banks don’t fail. First of all, banks are fundamentally owned by the government and when they make too many bad loans, they don’t typically fail. Usually the government, the regulators, come and somebody gets arrested and the government re-capitalizes the bank. It’s often very quiet – it’s not even necessarily announced to the world – and the bank keeps on going. What does that mean? That means that Chinese banks can take more risk than US banks can. In the US, we had almost no competitors because everybody thought “Lending to technology companies is way too risky, so we’ll just let Silicon Valley Bank do it. None of the rest of us will try.” In China, many, many, many banks want to copy us and do the same thing, because they’re not worried about what happens if we lose too much money. So that’s another big difference there…

…Wilcox: After I’d been there for several months, it occurred to me one day that my main conversation partner, the guy who is the Chairman, who was from Shanghai Pudong Development Bank, it occurred to me that he actually wears three hats. The only hat I wear is banker / businessman. But he had a banker / businessman hat, and he had a party hat, and he had a government hat. Then I started to wonder, when I’m talking with him, which hat is he wearing? It took me a long time before I figured out he doesn’t even think he has three hats. He thinks they’re all the same hat, so he’s not even thinking about it the same way I was. So I think that’s quite confusing. 

It’s also confusing when people find out, when a US company comes to China and finds out that it’s going to get a Party Committee in their organization. They get very confused because they don’t know what a Party Committee is. If you ask people in government or in the party, “What’s a Party Committee?” You say, “We’re going to have one , but I don’t understand what it is?” It’s hard for them to explain. You get multiple definitions and then you don’t know what is actually going to happen. Some people will tell me, “When you get a Party Committee, it’ll be so good because all the employees in your organization who are members of the party will have a place to gather once a month and discuss things.” Then somebody else says, “When you get a Party Committee, it’ll be so much easier because the Party Committee will help you put on social events for the employees, all of the employees.” But then somebody else told me, “No, when you get a Party Committee, it’ll be like another board, but a secret board. You won’t know who’s on it and they will influence what the real board does – or what I would call the real board.” Then other people told me, “Don’t pay any attention. That’s all silliness. There is no such thing as a Party Committee.” So it’s very, very confusing…

…Wilcox: I’ll give you the best example and that is that I believe based on the years I spent in China, that ultimately the main reason they wanted us in China – and they actually were very determined to get us to come to China. I remember that early on, a couple of years before my wife and I moved to China, I had a series of meetings with a very high-level government official who’s also got a lot of status in the party. He was saying to me, “Ken, we really want you to bring your bank to China. Your bank is more important than any bank we’ve ever met. You’re more important than – he explicitly said this – he says, You’re more important than Morgan Stanley and more important than Goldman Sachs. And by the way Ken, you’re one of the smartest Americans we’ve met.” So you think to yourself, “Well this is an exaggeration, but it does feel nice.” He obviously is going to help me get established in China. But what I didn’t realize is that the main reason they wanted us in China was so that they could study our business model and figure out how to copy it over time. That was something I wasn’t expecting, but I should have if I were less naive. If I were better prepared, I would have realized that was the intention. So the original title, the working title I had for my book, which I had to change because the publisher didn’t like it, my original title was, “One Bed, Two Dreams”, because that’s a phrase that most Chinese are familiar with. It explains why it didn’t work well, because my dream was working with all these Chinese technology companies and helping them do business with the rest of the world, and their dream was learning our business model.

The result was that when they gave us our license, they also told us that we would not be able to use Chinese currency for three years. That made it almost impossible to do business for the first three years. The people that said these things were both members of the government and members of the party. So I don’t know which one was talking. But they said, “We understand that you won’t be able to do much business for the first three years because the companies that you want to work with all want renminbi, they don’t want US dollars. But you can still be a good citizen. You can do what we would do, and that is we here in China help each other. So you can be helpful and prove that you care about China by teaching other banks your business model during the three years when you can’t really do much business. We’ll give you subsidy to help support you during the three years when you can’t earn much money because you can’t really do any business.” Then at the end of the three years when they gave us permission to use renminbi, they said to us, “We are so happy that you came to China and we really admire your business model and we admire it so much that we’re starting a bank of our own using your business model. Would you mind staying a little longer and being an advisor to this new bank that’s going to use your business model?” It felt like they were stealing my intellectual property but I’m not sure they thought of it that way…

…Wilcox: General Motors when it went over to China in 1985, the Chinese really didn’t have an auto industry. They wanted General Motors there not because they wanted General Motors to make a lot of money. It was because they wanted to learn about automobile manufacturing and because it took so long to build up the knowledge base, General Motors was welcome for about 30 years. But now General Motors is slowly losing market share and it’s probably going to withdraw from China. Then what will happen is China has made so much progress partially because they’re hardworking and smart, partially because they had General Motors there to learn from them, and then once General Motors retracts and goes back to the US, the auto industry in China will begin exporting and competing globally. I think actually the Chinese have done such a good job of first of all, learning from foreign automakers, but then on top of that, taking it further that the foreign automakers are in huge trouble. I think China’s automobile industry will dominate in the future. 

4. Weekend thoughts: crypto, mania, and reflexivity follow up – Andrew Walker

When I first saw the “BTC yield” metric, I thought it was pretty crazy. MSTR is trading for approaching 3x the value of their bitcoin; if they issue stock and use all the stock to buy bitcoin, of course it’s going to cause their bitcoin holdings per share to go up…. and even more so if they issue debt and use that to buy bitcoin and then judge themselves on a per share basis! Taken to its extreme2, if you thought BTC yield was truly the be all, end all of value creation, and the higher the BTC yield the better, then any company following a pure BTC yield strategy should lever themselves up to the maximum amount possible, no matter the terms, and use all of the proceeds to buy BTC. Obviously no one does that because it would be insanity and eventually banks would stop lending, but I illustrate that only to show that purely maximize BTC yield is clearly not value maximizing….

But, if you look at the fine print, BTC yield is even crazier than simply suggesting increasing BTC per share is the only value creation metric that matters. If you really look at the MSTR BTC yield table above or read their disclosures, you’ll notice that the BTC yield assumes that all of their convertible debt converts…

…So, go back to MSTR’s BTC yield table; they have a set of 2029 converts that convert at $672.40/share. Those are far, far out of the money (MSTR’s stock trades for ~$400/share as I write this)…. yet MSTR’s BTC yield assumes those converts are in the money / will convert for their BTC yield.

That is an insane assumption that casually assumes MSTR’s shares almost double3. And, again, by taking this assumption to its extreme, we can see how wild it is. Like all things, convert debt involves different trade offs; for example, you could get a higher strike price by taking on a higher interest rate (i.e. if your strike price is ~$670 at a 0% interest rate, you could probably push it up to $770 by taking on a 3% interest rate or $870 by taking on a 6% interest rate4). MSTR has issued all of this convert debt deals at 0% interest rates, which is a great pitch (“we’re borrowing for free, we don’t have to pay a carry to buy BTC, etc”)…. but if BTC yield is all that matters, MSTR could start issuing convertible debt with really high interest rates, which would jack that strike price of the convert up, thus decreasing dilution and increasing the BTC yield…

…MSTR fans would say “but raising converts with interest doesn’t make sense; it’s no longer free money / now it has a carry cost.” And I understand that argument…. but convertible debt isn’t free money either, and I just do this to highlight how insane BTC yield is as a be all / end all metric!…

…The BTC yield that all of these companies present assumes that their convert debt converts, and that is a big / crazy assumption…. but it’s interesting to think about what will happen in five years. There is, of course, a world where BTC goes to $250k (or higher) and all of these stocks moon. In that world, the converts will be well in the money, and all of this worry will sound silly…. but there is also a world where BTC stalls out or drops over the next few years, and that world is really interesting. All of these companies are raising converts with 5-7 year maturities, so if BTC doesn’t moon and the converts aren’t in the money, you’re going to have all of the BTC standard companies facing a maturity wall at the same time. What happens then? I doubt they can roll the converts at anything close to the same terms (remember, cheap converts require high volatility, and if the stocks have stalled out for five years vol is going to be a lot lower), so they’ll either need to sell a ton of equity to paydown the debt (which will be tough; there probably won’t be much enthusiasm for the stock, and I’m not sure the market would be able to absorb the hypothetical amount of stock they’d need to issue without some enthusiasm)…. or you’ll have a wave of BTC standard companies all looking to sell down some of their bitcoin to payoff converts at the exact same time.

5. Satya Nadella | BG2 (Transcript here)- Bill Gurley, Brad Gerstner, and Satya Nadella

Gerstner: Shifting maybe to enterprise AI, Satya. The Microsoft AI business has already reported to be about $10 billion. You’ve said that it’s all inference and that you’re not actually renting raw GPUs to others to train on, because your inference demand is so high. As we think about this, there’s a lot of skepticism out there in the world as to whether or not major workloads are moving. If you think about the key revenue products that people are using today and how it’s driving that inference revenue for you today, and how that may be similar or different from Amazon or Google, I’d be interested in that.

Nadella: The way for us this thing has played out is, you got to remember most of our training stuff with OpenAI is sort of more investment logic. It’s not in our quarterly results – it’s more in the other income, based on our investment.

Gerstner: Other income or loss right?

Nadella: That is right. That’s how it shows up. So most of the revenue or all the revenue is pretty much our API business or in fact, to your point, ChatGPT’s inference costs are there, so that’s a different piece. The fact is the big-hit apps of this era are ChatGPT, Co-Pilot, GitHub Co-Pilot, and the APIs of OpenAI and Azure OpenAI. In some sense, if you had to list out the 10 most hit apps, these would probably be in the four or five of them and so therefore that’s the biggest driver.

The advantage we have had, and OpenAI has had, is we’ve had two years of runway pretty much uncontested. To your point, Bill made the point about everybody’s awake and it might be. I don’t think there will be ever again maybe a two-year lead like this, who knows? It’s all you say that and somebody else drops some sample and suddenly blows the world away. But that said, I think it’s unlikely that that type of lead could be established with some foundation model. But we had that advantage, that was the great advantage we’ve had with OpenAI. OpenAI was able to really build out this escape velocity with ChatGPT.

But on the API side, the biggest thing that we were able to gain was.. Take Shopify or Stripe or Spotify. These were not customers of Azure, they were all customers of GCP or they were customers of AWS. So suddenly we got access to many, many more logos, who are all “digital natives” who are using Azure in some shape or fashion and so on. So that’s sort of one. When it comes to the traditional enterprise, I think it’s scaling. Literally it is people are playing with Co-Pilot on one end and then are building agents on the other end using Foundry. But these things are design wins and project wins and they’re slow, but they’re starting to scale. Again, the fact that we’ve had two years of runway on it, I think…

I like that business a lot more, and that’s one of the reasons why the adverse selection problems here would have been lots of tech startups all looking for their H100 allocations in small batches. Having watched what happened to Sun Microsystems in the dotcom, I always worry about that. You just can’t chase everybody building models. In fact, even the investor side, I think the sentiment is changing, which is now people are wanting to be more capital-light and build on top of other people’s models and so on and so forth. If that’s the case, everybody who was looking for H100 will not want to look for it more. So that’s what we’ve been selective on.

Gerstner: You’re saying for the others that training of those models and those model clusters was a much bigger part of their AI revenue versus yours? 

Nadella: I don’t know. This is where I’m speaking for other people’s results. It’s just I go back and say, “What are the other big-hit apps?” I don’t know what they are. What models do they run? Where do they run them? When I look at the DAU numbers of any of these AI products, there is ChatGPT, and then there is – even Gemini, I’m very surprised at the Gemini numbers, obviously I think it’ll grow because of all the inherent distribution. But it’s kind of interesting to say that they’re not that many. In fact, we talk a lot more about AI scale, but there is not that many hit apps. There is ChatGPT, Github Co-Pilot, there’s Co-Pilot, and there’s Gemini. I think those are the four I would say, in a DAU, is there anything else that comes to your mind?…

…Gurley: Satya, on the enterprise side, obviously the coding space is off to the races and you guys are doing well and there’s a lot of venture-backed players there. On some of the productivity apps, I have a question about the the Co-Pilot approach and I guess Marc Benioff’s been obnoxiously critical on this front, calling it Clippy 2 or whatever. Do you worry that someone might think first-principles AI from ground-up, and that some of the infrastructure, say in an Excel spreadsheet, isn’t necessary to know if you did a AI-first product. The same thing by the way could be said about the CRM right? There’s a bunch of fields and tasks that that may be able to be obfuscated for the user.

Nadella: It’s a very, very, very important question. The SaaS applications or biz apps, let me just speak of our own Dynamics thing. The approach at least we’re taking is, I think the notion that business applications exist, that’s probably where they’ll all collapse in the agent era. Because if you think about it, they are essentially CRUD databases with a bunch of business logic. The business logic is all going to these agents, and these agents are going to be multi-repo CRUD. They’re not going to discriminate between what the back-end is, they’re going to update multiple databases, and all the logic will be in the AI tier so to speak. Once the AI tier becomes the place where all the logic is, then people will start replacing the backends right? In fact it’s interesting, as we speak, I think we are seeing pretty high rates of wins on Dynamics backends and the agent use, an we are going to go pretty aggressively and try and collapse it all, whether it’s in customer service, whether it is in… 

By the way, the other fascinating thing that’s increasing is just not CRM, but even what we call finance and operations, because people want more AI-native biz app. That means the biz app, the logic tier, can be orchestrated by AI and AI agents. So in other words, Co-Pilot to agent to my business application should be very seamless.

Now in the same way, you could even say, “Why do I need Excel?” Interestingly enough, one of the most exciting things for me is Excel with Python, is like GitHub with Co-Pilot. So what we’ve done is, when you have Excel – by the way this would be fun for you guys – which is you should just bring up Excel, bring up Co-Pilot, and start playing with it. Because it’s no longer like – it is like having a data analyst, so it’s no longer just making sense of the numbers that you have. It will do the plan for you. It will literally – like how GitHub Co-Pilot Workspace creates the plan and then it executes the plan – this is like a data analyst who is using Excel as a sort of row/column visualization to do analysis scratch pad. So it kind of tools you. So the Co-Pilot is using Excel as a tool with all of its action space because it can generate and it has python interpreter. That is in fact a great way to reconceptualize Excel. At some point you could say, “I’ll generate all of Excel” and that is also true. After all, there’s a code interpreter, so therefore you can generate anything.\

So yes, I think there will be disruption. The way we are approaching, at least our M365 stuff is, one is build Co-Pilot as that organizing layer UI for AI, get all agents, including our own agents – you can say Excel is an agent to my Co-Pilot, Word is an agent, it’s kind of a specialized canvases, which is I’m doing a legal document, let me take it into Pages and then to Word and then have the Co-Pilot go with it, go into Excel and have the Co-Pilot go with it. That’s sort of a new way to think about the work in workflow…

…Gurley: Satya, there’s been a lot of talk about model scaling and obviously there was talk, historically about 10x-ing the cluster size that you might do, over and over again, not once and then twice. X.AI is still making noise about going in that direction. There was a podcast recently where they flipped everything on their head and they said “If we’re not doing that anymore, it’s way better because we can just move on to inference which is getting cheaper and you won’t have to spend all this capex. I’m curious, those are two views of the same coin. But what’s your view on LLM model scaling and training cost, and where we’re headed in the future?

Nadella: I’m a big believer in scaling laws I’ll first say. In fact, if anything, the bet we placed in 2019 was on scaling laws and I stay on that. In other words, don’t bet against scaling laws. But at the same time, let’s also be grounded on a couple of different things.

One is these exponentials on scaling laws will become harder, just because as the clusters become harder, the distributed computing problem of doing large scale training becomes harder. That’s one side of it. But I would just still say – and I’ll let the OpenAI folks speak for what they’re doing – but they are continuing to – pre-training I think is not over, it continues. But the exciting thing, which again OpenAI has talked about and Sam has talked about, is what they’ve done with o1. This Chain of Thought with autograding is just a fantastic. In fact, basically, it is test-time compute or inference-time compute as an another scaling law. You have pre-training, and then you have effectively this test-time sampling that then creates the tokens that can go back into pre-training, creating even more powerful models that then are running on your inference. So therefore, that’s I think a fantastic way to increase model capability.

The good news of test-time or inference-time compute is sometimes, running of those o1 models means… There’s two separate things. Sampling is like training, when you’re using it to generate tokens for your pre-training. But also customers, when they are using o1, they’re using more of your meters, so you are getting paid for it. Therefore, there is more of an economic model, so I like it. In fact, that’s where I said I have a good structural position with 60-plus data centers all over the world.

Gurley: It’s a different hardware architecture for one of those scaling versus the other, for the pre-training versus…

Nadella: Exactly. I think the best way to think about it is, it’s a ratio. Going back to Brad’s thing about ROIC, this is where I think you have to really establish a stable state. In fact, whenever I’ve talked to Jensen, I think he’s got it right, which is you want to buy some every year. Think about it, when you depreciate something over 6 years, the best way is what we have always done, which is you buy a little every year and you age it, you age it, you age it. You use the leading node for training and then the next year it goes into inference, and that’s sort of the stable state I think we will get into across the fleet for both utilization and the ROIC and then the demand meets supply.

Basically, to your point about everybody saying, “Have the exponentials stopped?” One of the other things is the economic realities will also stop, right? At some point everybody will look and say, “What’s the economically rational thing to do?” Which is, “Even if I double every year’s capability but I’m not able to sell that inventory,” and the other problem is the Winner’s Curse, which is – you don’t even have to publish a paper, the other folks have to just look at your capability and do either a distillation… It’s like piracy. You can sign off all kinds of terms of use, but like it’s impossible to control distillation. That’s one. Second thing is, you don’t even have to do anything, you just have to reverse engineer that capability and you do it in a more computer efficient way. So given all this, I think there will be a governor on how much people will chase. Right now a little bit of everybody wants to be first. It’s great, but at some point all the economic reality will set in on everyone and the network effects are at the app layer, so why would I want to spend a lot on some model capability with the network effects are all on the app?…

…Gurley: Does your answer to Brad’s question about the balancing of GPU ROI, does that answer the question as to why you’ve outsourced some of the infrastructure to Coreweave in that partnership that you have?

Nadella: That we did because we all got caught with the hit called ChatGPT. It was impossible. There’s no supply chain planning I could have done. None of us knew what was going to happen. What happened in November of ‘22, that was just a bolt from the blue, therefore we had to catch up. So we said, “We’re not going to worry about too much inefficiency.” That’s why whether it’s Coreweave or many others – we bought all over the place. That is a one time thing and then now it’s all catching up. That was just more about trying to get caught up with demand.

Gerstner: Are you still supply-constrained Satya?

Nadella: Power, yes. I am not chip supply-constrained. We were definitely constrained in ‘24. What we have told the street is that’s why we are optimistic about the first half of ‘25, which is the rest of our fiscal year and then after that I think we’ll be in better shape going into ‘26 and so on. We have good line of sight.

---

Disclaimer: The Good Investors is the personal investing blog of two simple guys who are passionate about educating Singaporeans about stock market investing. By using this Site, you specifically agree that none of the information provided constitutes financial, investment, or other professional advice. It is only intended to provide education. Speak with a professional before making important decisions about your money, your professional life, or even your personal life. We currently have a vested interest in Alphabet (parent of Google), Amazon, Meta Platforms (parent of Facebook), and Microsoft. Holdings are subject to change at any time.

We’ve constantly been sharing a list of our recent reads in our weekly emails for The Good Investors.

Do subscribe for our weekly updates through the orange box in the blog (it’s on the side if you’re using a computer, and all the way at the bottom if you’re using mobile) – it’s free!

But since our readership-audience for The Good Investors is wider than our subscriber base, we think sharing the reading list regularly on the blog itself can benefit even more people. The articles we share touch on a wide range of topics, including investing, business, and the world in general. 

Here are the articles for the week ending 15 December 2024:

1. SpaceX: Rocket Ship – Matt Reustle and Luke Ward

Luke

So if we take the CapEx part of that first, NASA estimated that the cost to develop the Falcon 9 from scratch would be about $4 billion. But SpaceX ended up doing it for about a tenth of that price. So to begin with, that’s an order of magnitude improvement in the level of investment required.

SpaceX gives you the prices for launches on their website. So $70 million per launch of a Falcon 9 flight—that’s already 20 times cheaper than the Space Shuttle was per kilogram into orbit. But the real kicker, as you point out, is the operating leverage that comes from having partial reusability…

…Starship is designed to be fully and rapidly reusable. So unlike Falcon 9, which is only partially reusable but also able to fly multiple times every day, it’s going to have a payload capacity that’s about 100 tons to orbit at the beginning, but probably rising to closer to 200 tons to orbit over time.

And Musk has suggested that a variable cost of around $10 million per launch is the ballpark figure which they’d be aiming for at scale in a steady state, ambitiously maybe even falling to $2 million—a figure which has been touted. If you believe those kinds of performance levels are feasible, that gets the cost down to around $10 per kilogram. That’s over 100 times cheaper than the Falcon 9 we’re talking about at the moment. And that would have a dramatic effect on what’s economically feasible for humanity to do in space…

…Matt

Satellites in Low Earth Orbit—there is quite a bit of history in terms of that being the obvious space use case, that having an existing economy. I think Starlink is an extension of that. Different, absolutely, but an extension of what was going on.

Are there brand new industries being unlocked or obvious things with line of sight that open up from a space economy perspective that you see either today or, when I say near future, you could extend that out however far you think is reasonable.

Luke

A lot of these options which SpaceX has to develop, brand new markets that don’t exist already, are a function ultimately of the cost curve. Take semiconductor manufacturing on Earth; at the moment, we spend billions of dollars per fab to recreate the conditions which are readily accessible in space for free, if you can get there.

And so there’s some point on the cost curve intersecting between the cost of building a fab and the cost of launching a fab or the equipment of a fab into orbit and operating there instead. Same can be said of pharmaceutical research. The crystallization structures which are able to happen in space are different from the ones which are able to happen under the influence of gravity.

So if you think about pricing on pharmaceuticals, extending patent lives, etc., if you can move the manufacturing or the research lab for cutting-edge pharmaceuticals into space, you could make high-value, low-volume products. Something which would really make sense to do and doesn’t require a huge technological innovation to happen.

The list can go on and on—artificial organs, for example, being able to manufacture perfectly spherical lenses. There’s lots and lots of things which could be made.

Maybe the way to think about that is that space-based manufacturing could be the next large market for this if the costs can continue to come down. Starship having the volume of an A380 or a 747—think of the equivalent size of factory that represents. And if that can be launched every single day and recovered every single day for $10 per kilogram, that could be a really compelling way to do quite a lot of manufacturing.

Incidentally, that’s something that Jeff Bezos really focuses on in his vision for space as opposed to Mars per se, is where we can move a lot of the heavy-polluting industry off the planet. And why don’t we turn Earth into this perfect nature reserve, and all these polluting aspects of manufacturing can go into orbit, which again is very compelling.

Probably needs a lot more innovation to deliver communications from orbit, but I’d say it’s maybe an inevitability if the cost gets to a low enough point. You think how much solar energy is available without the atmospheric attenuation, for example—you know, 24/7. There’s lots of compelling reasons why if it’s cheap enough, at some point a lot of these things probably should happen, not just could happen.

Matt

The solar energy point, great example of something that is an entirely different dynamic in space than on Earth. What would the other things be? Just out of curiosity, when you mentioned semiconductors or pharmaceuticals, is it just purely gravity? Are there other things that are happening in space or not happening in space that happen on Earth that would drive that difference?

Luke

There’s the vacuum conditions—so there isn’t an atmosphere—so the level of impurities which you need to get rid of for a vapor deposition machine, for example. You don’t have the same kind of challenges there of having to have this deep vacuum.

Then, arguably, in space, because you don’t have gravity, you could construct much larger structures there rather than construct them on the ground and then launch them.

So again, that volume constraint which we were talking about earlier, in terms of how big your payload is—if you’re able to get enough stuff up there and assemble it in space, as we did with the International Space Station, things can be much, much larger given the payload bay of Starship than they could with the Space Shuttle.

Matt

When you think about low Earth orbit versus geosynchronous orbit versus something like Mars—which I think was the original vision with Elon and SpaceX—how much does that change the economics when you extend out?

Is it orders of magnitude where it’s an exponential cost curve to go further out? Even just if we focus on the launch and use a satellite for an example, before we get into all the manufacturing dynamics, is there any way to contextualize that from a cost perspective?

Luke

The really good news here is that gravitational force decreases with the square of distance. So the biggest challenge is getting off the surface and into orbit. Once you’re there, from an energy point of view, it’s a lot easier to go anywhere else in the solar system.

So if you were to take Falcon 9 again as the example, for the same price, it can place 20 tons into low Earth orbit, or it can place 4 tons into Martian orbit. That’s despite the latter being over a million times further away. Now, this feeds into what I think is probably the biggest misconception about SpaceX and its Mars ambitions.

I’d say for most people, the idea of a commercial entity pursuing exploration is naive at best. But I’d argue that long-term investors should be absolutely ecstatic about SpaceX having this mission as a forcing function. Firstly, it’s the key to getting the best people in the world to come and work for the organization and allow it to innovate in a manner and speed that others simply can’t match. That’s a huge competitive advantage.

Secondly, the way to get more cargo to Mars is actually about figuring out how to get more cargo into orbit around Earth, because that’s where the cost is all concentrated. It’s all in that first initial leap off the surface of our planet. So rather than framing Starship as a system that makes it possible to get to other planets, think about it instead being a system that could make it enormously more profitable to operate a business in Earth orbit and unlock brand new commercial use cases there as well…

…Luke

When we talk to SpaceX, they’re still very much focused on the here and now in the next couple of years. They have ambitions for things which they could do, but the focus is very much on the core business: serving the core customers, serving Starlink, getting Starship to launch status. We’ll deal with the next things next.

They’ve got so many things which they could be doing at the moment. When we come to this, a lot of this is us hypothesizing of how that could evolve beyond information which they’ve given us. The trend which you’ve seen of them to be vertical integrators could be quite informative. It might be that they end up being the ones who are commercializing a lot of these other services.

Rather than having a customer paying them for it at substantial scale, it would make more sense for them to do it. Could you start seeing some of these aspects? If they get into space-based manufacturing, for example, could that be priced on a value-added basis rather than a subscription basis or a volume basis? Certainly seems possible. If you start running data centers in space because it’s easier to power or cool them, etc., could you start offering data storage and machine learning alongside Starlink connectivity?

The further you look out, the more and more wacky it can get, but it’s also potentially financially plausible as well. You maybe have to take a bit of inspiration from science fiction here, but it’s quite a common trope in some of these movies of these large mega-corporations—the Weyland-Yutani Corporation from the Alien movies, or the Resources Development Administration from the Avatar films—where one mega-corporation was able to dominate access to space early on and then ends up controlling the entire extrasolar economy because of the advantages it had at that really early stage…

…Luke

The human spaceflight at the moment definitely has been the preserve of the rich and famous, but at scale that becomes cheaper and cheaper. And if we are talking about launching, Starship could be used as much for sending cargo and people to other points on the planet rather than other points in space. And so one option that the government’s looking into is this notion of rocket cargo delivery. Starship would be able to deliver 200,000 kg anywhere on the planet within 40 minutes.

What does that do for sort of a rapid reaction force, and what does that do for next-day delivery? At some stage, it’s going to be feasible to put a lot of astronauts or paying passengers on something like that, and it will be a quicker and potentially more efficient way to do long-distance travel. These things really could get quite wild, but it could be plausible at some stage. Again, that’s not the reason to invest in the company today; that’s not the basis of what they’re doing, and it’s a lot of people getting excited about things.

But come back in 10 years, I’d be disappointed if you or I weren’t able to go into space at some point in our lifetime for the cost of a premium economy ticket or something like that.

2. Japan vs Big Tech – Daye Deng

Put simply, US big tech has grown so dominant that it’s singlehandedly blowing a hole in the trade balance of a nation as large as Japan…

…In 2023, Japan recorded JPY 5.5 trillion in so-called digital trade deficit. The Ministry of International Trade and Industry (MITI) projects this to grow to JPY 8 trillion by 2030, at which point it could surpass Japan’s annual import of crude oil.

Japan’s total goods and services trade deficit in 2023 was JPY 6 trillion, with the digital deficit accounting for JPY 5.5 trillion…

…Japan has been in a structural deficit for goods trade over the past two decades. This may come as a surprise to those who have held onto the old idea that Japan is an export powerhouse.

There are several reasons for the shift:

• Japanese firms have moved production overseas. This isn’t entirely negative since Japanese firms (and their profits) continue to grow, but it has contributed to a widening trade deficit.
• Japan’s loss of global competitiveness in certain industries, like chips and appliances, to rivals such as South Korea.
• Rising cost of imports driven by energy shocks, rising overseas inflation, and weak yen.

The third point deserves elaboration. Japan’s reliance on imported energy has long been a critical structural weakness. For example, following 2011 Fukushima nuclear disaster, Japan significantly reduced domestic nuclear energy production and increased its reliance on imported LNG, becoming a major contributor to trade deficit.

A similar pattern emerged post-Covid. Global oil and commodity prices surged. This was compounded by high rates of overseas inflation on general imports. On top, a historically weak yen made imports even more expensive…

…Since 2014, the Japanese government has been disclosing the digital deficit, which has grown 2.6-fold from 2014 to JPY 5.5 trillion in 2023. This is a net figure derived from JPY 9.2 trillion paid for digital services and JPY 3.7 trillion received from abroad…

…The picture is quite clear: on the services side, Japan is taking its hard-earned surplus from tourism and spending it all on paying for digital services.

How will this play out? While I’m personally bullish on the Japanese tourism industry, it still has natural growth constraints. However, there is no ceiling on how much Japan can continue to spend on digital services. In fact, digital services spend could accelerate given:

• Japan is already playing catch-up in the digital realm, and is behind other major countries in many key digital metrics.
• AI is poised to make Japan’s digital dependency crisis even worse, in a world where firms like Nvidia and those that are able to scale AI services (e.g. hyperscalers) dominate AI economics.

Without an AI champion of its own, Japan has few options if it wants to avoid being left behind in the new digital paradigm…

…Based on our discussion so far, does it surprise you that the Japanese yen has been weak?

“According to an analysis by Mizuho Research & Technologies, if the digital deficit doubles from the 2023 level by the end of March 2026, it will add another 5 to 6 yen of depreciation in the Japanese currency’s value against the dollar.”

– Nikkei Asian Review

Or let me put it another way — would you feel bullish about the currency of a country that relies on tourism as its primary growing surplus, while ultimately funneling all those earnings (and more) into paying for essential energy imports and ever-increasing digital spend on big tech?…

…In recent years we’ve seen how hard Japan has been trying to reclaim its position in the semiconductor industry. But do they only care about hardware and not its digital sovereignty? Will Japan continue to sit back and let US tech giants profit endlessly, or will it finally confront its position as a digital colony?

3. Guyana and the mystery of the largest ranch in the Americas – Swen Lorenz

Many mistakenly believe that Guyana is located in Africa – when it’s actually nestled right next to Venezuela…

…In 2015, ExxonMobil discovered oil off the coast of Guyana.

The discovery changed the course of the country. Long one of the poorest nations of the Western hemisphere, Guyana has since become the world’s fastest growing economy.

Since 2015, its GDP per capita has more than quintupled. In 2022 and 2023, its economy grew by 67% and 33%, respectively. Another stunner of a year is forecast for 2024, with 34% GDP growth.

The former British colony benefits from a large amount of oil wealth spread around a relatively small population of 800,000 people. Per head, there is twice as much oil as in Saudi Arabia. To put things in perspective, Guyana’s landmass is nearly as big as the UK, but it only has 1.2% of the UK’s population…

…Just a week ago, ExxonMobil reported that it had reached 500m barrels of oil produced in Guyana since output began in 2019. The goal is to lift production to 1.3m barrels per day by 2027, up from currently 650,000 barrels. In comparison, the UK’s North Sea produces just 1m barrels per day…

…Supporters of the country’s energy projects claim that they will bring untold riches to the population. Indeed, Guyana recently started to hand out cheques to its citizens, including the Guyanese diaspora of 400,000 people, who the government encourages to come back as it needs more labour to support the strong economic growth.

4. Capital, Compute & AI Scaling – Patrick O’Shaughnessy, Chetan Puttagunta, and Modest Proposal

Modest

Everyone knows the Mag 7 represent a larger percent of the S&P 500 today. But beyond that, I think thematically AI has permeated far broader into industrials, into utilities and really makes up, I would argue, somewhere between 40 and 45% of the market cap as a direct play on this. And if you even abstract to the rest of the world, you start bringing in ASML, you bring in TSMC, you bring in the entire Japanese chip sector. And so if you look at the cumulative market cap that is a direct play on artificial intelligence right now, it’s enormous…

… I think at the micro level this is a really powerful shift if we move from pre-training to inference time and there are a couple big ramifications.

One, it better aligns revenue generation and expenditures. I think that is a really, really beneficial outcome for the industry at large, which is in the pre-training world you were going to spend 20, 30, $40 billion on CapEx, train the model over 9 to 12 months, do post-training, then roll it out, then hope to generate revenue off of that in inference. In a test time compute scaling world you are now aligning your expenditures with the underlying usage of the model. So just from a pure efficiency and scalability on a financial side, this is much, much better for the hyperscalers.

I think a second big implication, again we have to say we don’t know that pre-training scaling is going to stop. But if you do see this shift towards inference time, I think that you need to start to think about how do you re-architect the network design? Do you need million chip super clusters in energy low-cost land locations or do you need smaller, lower-latency, more efficient inference-time data centers scattered throughout the country? And as you re-architect the network, the implications on power utilization, grid design?

A lot of the, I would say, narratives that have underpinned huge swaths of the investment world I think have to be rethought and I would say today because this is a relatively new phenomenon, I don’t believe that the public markets have started to grapple with what that potential new architecture looks like and how that may impact some of the underlying spend…

… Chetan

But at the moment, at this plateauing time, we’re starting to see these small teams catch up to the frontier. And what I mean by frontier is where are the state-of-the-art models, especially around text performing? We’re seeing these small teams of quite literally two to five people jumping to the frontier with spend that is not one order, but multiple orders of magnitude less than what these large labs were spending to get there.

I think part of what’s happened is the incredible proliferation of open-source models. Specifically, what Meta’s been doing with LLaMA has been an extraordinary force here. LLaMA 3.1 comes in three flavors, 405 billion, 70 billion, 8 billion. And then LLaMA 3.2 comes in 1 billion, 3 billion, 11 billion, and 90 billion.

And you can take these models, download them, put them on a local machine, you can put them in a cloud, you can put them on a server, and you can use these models to distill, fine-tune, train on top of, modify, et cetera, et cetera, and catch up to the frontier with pretty interesting algorithmic techniques.

And because you don’t need massive amounts of compute, or you don’t need massive amounts of data, you could be particularly clever and innovative about a specific vertical space, or a specific technique, or a particular use case to jump to the frontier very, very quickly…

…Chetan

The force of Llama today has been two things, and I think this has been very beneficial to Meta is one. The transformer architecture that Llama is using is a sort of standard architecture, but it has its own nuances.

And if the entire developer ecosystem that’s building on top of Llama is starting to just assume that that Llama 3 transformer architecture is the foundational and sort of standard way of doing things, it’s sort of standardizing the entire stack towards this Llama way of thinking, all the way from how the hardware vendors will support your training runs to the hyperscalers and on and on and on. And so standardizing on Llama itself is starting to become more and more prevalent.

And so if you were to start a new model company, what ends up happening is starting with Llama today is not only great because Llama is open source, it’s also extraordinarily efficient because the entire ecosystem is standardizing on that architecture…

…Modest

So I think the interesting part for OpenAI was because they just raised the recent round and there was some fairly public commentary around what the investment case was. You’re right, a lot of it oriented around the idea that they had escape velocity on the consumer side and that ChatGPT was now the cognitive reference and that over time they would be able to aggregate an enormous consumer demand side and charge appropriately for that and that it was much less a play on the enterprise API and application building.

And that’s super interesting if you actually play out what we’ve talked about when you look at their financials, if you take out training runs, if you take out the need for this massive upfront expenditure, this actually becomes a wildly profitable company quite quickly in their projections. And so in a sense it could be better.

Now then the question becomes what’s the defensibility of a company that is no longer step function advancing on the frontier?…

…Chetan

These products are truly, as a software investor, absolutely amazing.

They require a total rethinking from first principles on how these things are architected. You need unified data layers, you need new infrastructure, you need new UI and all this kind of stuff. And it’s clear that the startups are significantly advantaged against incumbent software vendors. And it’s not that the incumbent software vendors are standing still, it’s just that innovator’s dilemma in enterprise software is playing out much more aggressively in front of our eyes today than it is in consumer.

I think in consumer, the consumer players recognize it, are moving it, and are doing stuff about it. Whereas I think in enterprise, even if you recognize it, even if you have the desire to do something, the solutions are just not built in a way that is responsive to dramatic re-architecture. Now could we see this happening? Could a giant SaaS company just pause selling for two years and completely re-architect their application stack?

Sure, but I just don’t see that happening. And so if you just look at any sort of analysis on what’s happening on AI software spend, something like it’s 8x year-over-year growth between 2023 and 2024 on just pure spend. It’s gone from a couple of hundred million dollars to well over a billion in just a year’s time…

…Modest

If you listen to AWS, one of the fascinating things they say is they call AWS a logistics business.

I don’t think anyone externally would sort of look at cloud computing and say, oh yeah, that’s a logistics business. But their point is essentially what they have to do is they have to forecast demand and they have to build supply on a multi-year basis to accommodate it.

And over 20 years they’ve gotten extraordinarily good at what has happened in the last two years, and I talked about this last time, is you have had an enormous surge in demand hitting inelastic supply because you can’t build data center capacity in three weeks. And so if you get back to a more predictable cadence of demand where they can look at it and say, okay, we know now where the revenue generation is coming from.

It’s coming from test time, it’s coming from Chetan and his companies rolling out. Now we know how to align supply with that. Now it’s back to a logistics business. Now it’s not grab every mothballed nuclear site in the country and try to bring it online.

And so instead of this land grab, I think you get a more reasonable, sensible, methodical rollout of it maybe. And I actually would guess that if this path is right, that inference overtakes training much faster than we thought and gets much bigger than we may have suspected.

But I think the path there in the network design is going to look very different and it’s going to have very big ramifications for the people who were building the network, who were powering the network, who were sending the optical signals through the network. And all of that, I think, has not really started to come up in the probability-weighted distributions of a huge chunk of the public market.

And look, I think most people overly fixate on NVIDIA because they are sort of the poster child of this, but there are a lot of people downstream from NVIDIA that will probably suffer more because they have inferior businesses. NVIDIA is a wonderful business doing wonderful things. They just happen to have seen the largest surge in surplus. I think that there are ramifications far, far beyond who is making the bleeding edge GPU, even though I do think there will be questions about, okay, does this new paradigm of test time compute allow for customization at the chip level much more than it would have if we were only scaling on pre-train…

…Modest

If you think about a training exercise, you’re trying to utilize them at the highest possible percent for a long period of time. So you’re trying to put 50, 100,000 chips in a single location and utilize them at the highest rate possible for nine months. What’s left behind is a hundred thousand chip cluster that if you were to repurpose for inferencing is arguably not the most efficient build because inference is peaky and bursty and not consistent.

And so this is what I’m talking about that I just think from first principles you are going to rethink how you want to build your infrastructure to service a much more inference focused world than a training focused world. And Jensen has talked about the beauty of NVIDIA is that you leave behind this in place infrastructure that can then be utilized.

And in a sunk cost world you say, sure, of course if I’m forced to build a million chip supercluster in order to train a $50 billion model, I might as well sweat the asset when I’m done. But from first principles it seems clear you would never build a 350,000 chip cluster with 2 1/2 gigawatts of power in order to service the type of request that Chetan’s talking about.

And so if you end up with much more edge computing with low latency and high efficiency, what does that mean for optical networking? What does that mean for the grid? What does that mean for the need for on site power versus the ability to draw from the local utility?…

…Chetan

Semiconductor company called Cerebras, and they recently announced that inference on Llama 3.1 405 billion for Cerebras is it can generate 900-plus tokens per second, which is a dramatic order-of-magnitude increase. I think it’s like 70 or 75 times faster than GPUs for inference as an example. And so as we move to the inference world, the semiconductor layer, the networking layer, et cetera, there’s tons of opportunities for startups to really differentiate themselves…

…Modest

On a less sort of dramatic view, the way I think about this, there’s AlphaGo, which famously did that move that no one had ever seen, and I think it’s like move 37, everybody was super confused about, ended up winning. And another example I love is Noam Brown, because I like poker, talked about his poker bot confused—it was playing high stakes, no limit, and it continually over-bet dramatically larger sizes than pros had ever seen before.

And he thought the bot was making a mistake. And ultimately it destabilized the pros so much. Think about that. A computer destabilized humans in their approach that they have to some extent taken on over-betting now into their game.

And so those are two examples where if we think about pre-training being bounded by the data set that we’ve given it, if we don’t have synthetic data generation capabilities, here you have two examples where algorithms did something outside of the bounds of human knowledge. And that’s what’s always been confusing to me about this idea that LLMs on their own could get to superintelligence, is functionally they’re bounded by the amount of data we give them up front.

5. Will China Take Over the Global Auto Industry? – Brad Setser

China has, according to the New York Times, the capacity to produce over 40 million internal combustion engine (ICE) cars a year.

Goldman Sachs thinks China will also have the capacity to produce around 20 million electric vehicles by the end of 2024…

…China’s internal market is around 25 million cars, and not really growing —so rising domestic EV sales progressively frees up internal combustion engine capacity for export.   Domestic demand for traditional cars is likely to be well under 10 million cars next year given the enormous shift toward EVs now underway inside China…

…Historically, the autos market has been largely regional (setting aside trade in luxury cars, where volumes are smaller). Most cars sold in China were made in China, most cars sold in Europe are produced in Europe, most cars sold in the North America are produced in North America, and so on. The U.S. did import a few million cars, on net, from Asia, and China imported a million or so luxury cars from Europe, but those were the exceptions rather than the rule.

That could change, absent hefty restrictions on Chinese auto imports (like the 100 percent tariff the U.S. now levies on EVs imported from China).

The global market—with massive overcapacity in China’s internal combustion engine (ICE) sector, massive capacity expansion in China’s EV sector, effectively unlimited credit for Chinese manufacturing firms from China’s state banks, and a Chinese yuan that is weaker against the dollar than it was back in 2008—is pushing for global auto manufacturing to become more like global electronics manufacturing, with a concentration of global production in a single region and, for that matter, a single country…

…Overcapacity in China’s automotive sector is not, in fact, all that new.

China’s traditional automotive sector was dominated by the joint ventures (“JVs”) formed by the large foreign firms and their (typically state-owned) Chinese partners. Chinese auto demand took off after the global financial crisis, and global firms responded by massively expanding their Chinese production capacity – as only the German luxury markets were interested in paying the 25 percent tariff and supplying the Chinese market from abroad.

But demand growth eventually slowed, and by 2018, the Wall Street Journal was reporting that the Chinese market was oversupplied…

…China’s EV industry—like EV industries in the U.S. and Europe—initially received substantial state backing. Chinese EV manufactures benefitted from downstream subsidies that built out China’s battery and battery chemical industry, as well as access to the world’s cheapest steel.
EV firms benefitted from cheap state financing—both equity injections from a myriad of state-backed funds and loans from state banks who (still) have to meeting lending quotas.

Moreover, China was quite explicitly protectionist in the application of its “consumer” EV subsidies.

Only EVs that were on state lists of qualifying vehicles were eligible for the subsidy, and the subsidy was only provided to cars that were made in China…

…And initially, only cars that were made in China with a battery made in China by a Chinese firm qualified for the lists…

…The only exception to the basic rule that qualifying for the list required using a battery made in China by a Chinese firm only confirmed the broad pattern of discrimination: Chinese-owned Volvo was allowed to use a Korean battery in one of its early EVs.

State support has not disappeared in any way as China’s EV industry took off.   Looking at direct cash subsidies from the central government to the manufacturers misses the myriad of ways China, Inc helps out firms producing in China…

…Nio received a significant ($1.9 billion) equity investment from the City of Hefei and the Province of Anhui, helping to offset ongoing losses. That equity injection was on top of state support for a factory in Hefei, which The New York Times reports was effectively a gift from the local government.

“‘The local government provided the land and the building’, said Ji Huaqiang, Nio’s vice president for manufacturing. ‘Nio does not own the factory or the land — it is renting, but the factory was custom built for Nio’”

That kind of support explains how Nio managed to build out its EV capacity even when its existing factories weren’t really being used that much:

“Nio’s two factories give it the capacity to assemble 600,000 cars a year, even though its annual rate of sales this autumn [2023] is only about 200,000 cars. Nio is nonetheless already building a third plant.”…

...What’s even more striking is that the investments that built out China’s EV capacity came in a market that was already saturated with modern auto production capacity.  That kind of investment wouldn’t have taken place without state guidance and support, support that was intended both to develop an indigenous Chinese industry (See China 2025) and to support a green transition that would reduce Chinese dependence on import fossil energy. It was the result of policy driven by the central government and backed financially by all levels of government. It also worked, China is now the world leader in EVs and batteries…

…If the world’s global firms can only compete with Chinese firms by using Chinese batteries and Chinese parts, that will hollow out much of the automotive industries of Europe and North America—a European brand on a Chinese-made car with a Chinese battery and drive train won’t sustain the current European auto supply chain or current European employment in the auto industry.

---

Disclaimer: The Good Investors is the personal investing blog of two simple guys who are passionate about educating Singaporeans about stock market investing. By using this Site, you specifically agree that none of the information provided constitutes financial, investment, or other professional advice. It is only intended to provide education. Speak with a professional before making important decisions about your money, your professional life, or even your personal life. We currently have a vested interest in ASML, Meta, and TSMC. Holdings are subject to change at any time.