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 14 June 2026:
1. Gas Prices, Stock Bubbles, Grad Advice — And Teaching Personal Finance In School (Transcript here) – Morgan Housel
I want to start with what is the biggest economic news story of this year: the war in Iran. For most of you listening or watching, the biggest impact that’s had on your life is the rise in gas prices and oil prices. I want to make a very nuanced point here about making predictions about the future, which is so common in economics, and so difficult and humbling.
When the war in Iran first started about three months ago, it was very common among the smartest, most astute, most educated economists, oil analysts, and talking heads to make predictions along these lines: if the Strait of Hormuz is closed for another week or two, you’re not going to see a rise in oil prices — you’re going to see an explosion of oil prices. Not $100 a barrel, but $150, $200, $250. Not $4 gas, but $7, $8, $9 gas, with flights being cancelled. Those predictions have been made for months, and it was always along the lines of “if it stays closed for another week or two, this is going to happen.”
I want to make this point without minimizing what’s happened to gas prices all over the world and what could happen in the future. I don’t want to say, “Look at all these people — they were wrong,” because the price of oil today is about where it was three months ago when the war first started. It surged and then plateaued at this level. That is something almost no one watching this three months ago would have predicted. Virtually everybody, if you had told them we would be three months into this war with the Strait of Hormuz closed, would have said this is going to be a Mad Max scenario in oil. And so far, as I record this, it has not been.
I want to make an important point here without making any predictions about what might happen next — almost the opposite point, about why these kinds of things happen. There is such a long history in economics, in politics, and in any kind of social world that makes predicting what’s going to happen next so hard, even when it seems like the most rational conclusion. It’s so appealing and so easy to make simple predictions: if X happens, Y will be the result. Very appealing, and I think very comforting, because when you make a prediction like that, it gives you — or the person listening to that forecast — a sense of control in a world that is uncertain, if not unpredictable.
I say this with the glory of hindsight and nothing else; I would not have known any of this three months ago. But from my understanding, a lot of why oil has not yet reached those Mad Max levels, despite being three months into the Strait of Hormuz closure, comes down to a few reasons. Number one, the United States is exporting oil and gas like never before, which has taken some of the supply-crunch pressure off. Number two, Saudi Arabia has a series of oil pipelines that have been massively extended and expanded over the last three months — one big pipeline going to the Red Sea has gone from 2 million barrels a day to 7 million barrels a day, taking a lot of pressure off oil that used to go through the Strait of Hormuz. Number three, China has massively decreased the level of its oil imports. And number four, all over the world, we’ve been draining down oil stocks and reserves. Can that last forever? Of course not. I’m not making any predictions about what’s going to happen next.
The point I want to make — in a much broader way that applies to so many more things in the world of money and economics than just the Strait of Hormuz and oil prices — is that it is extremely difficult to know how people are going to adapt and evolve to a change in the economy. It’s very easy to say “if X, then Y,” and it makes a lot of sense and it’s very comforting. It’s much more difficult to say, “If the Strait of Hormuz closes, then people are going to adapt in this way, and this way, and this other way, and therefore we don’t really know what the end result is going to be.” There are so many cases like this, whether it’s housing prices, stock prices, whatever it might be.
I’ll give you one example that was crazy at the time. 2009 was one of the worst years in economic history since the Great Depression — absolutely dreadful, during the financial crisis. Stocks finished up that year. They increased. It’s so easy to say that the second coming of the Great Depression would be bad for the stock market; that’s a very easy prediction to make. It was much more difficult to see how people, prices, and valuations would adapt and respond in that era.
I was thinking about this recently because gas prices in my town went up tremendously in the last three months, but they’ve been about the same for the last two and a half. They exploded at the beginning of the war and then plateaued. Looking into how the global oil market has adapted and evolved — and again, maybe that doesn’t last forever, I’m not making any prediction — it’s so important to have a sense of humility about how complex the global economy is, and about people’s ability to adapt in ways you never saw coming. That’s what makes predictions about what’s going to happen this year, next year, or over the next month so difficult.
The last thing I’ll say about the psychology of making predictions: it is very common that the higher the stakes, the more people are willing to believe forecasts. When the stakes are really high — gas prices could explode so much that you can’t afford your commute, or your flights are cancelled — people are willing to believe anybody who says, “I can tell you what’s going to happen next.” That becomes very appealing. The irony is that when the stakes are that high and things are moving that quickly, that’s when forecasts become the least reliable. The demand for forecasts increases exactly when the forecasts themselves become least reliable, because people are adapting and changing so quickly. That is why there is such a long history of economic forecasts for things that never happened.
2. Avoiding Death on the Yellow Brick Road – Joe Schmidt IV
The Yellow Brick Road is our shorthand for the path the labs are walking, where they’re committing extraordinary resources. The reason the labs are best-suited for problems like code generation, writing, or image-creation is because these problems improve with raw model capability: every dollar spent on pre-training and post-training improves product quality. Meanwhile, the rest of Oz is inhabited by more complex, often vertical problems, that aren’t as simple as giving a business user a horizontal tool with access to standard tools and computer use. The value comes less from the underlying model’s raw capability (though that’s still important!) than from the scaffolding around it that makes the output trustworthy, compliant, and operational inside a specific industry…
…The labs will certainly improve, but I’d argue there are a few ways the rest of Oz can defend themselves over time:
Data and learning flywheels: A lot of what you internalize isn’t in any training set — unwritten industry norms, undocumented standards, the tribal knowledge that lives in practitioners’ heads. None of it is on the public web. No amount of training compute substitutes for being inside the workflows where this knowledge actually lives. There are two flywheels stacked on top of each other here: an across-customer one — patterns that compound as you see more variants of the same problem — and a within-customer one — the why behind specific decisions, the unsaid exceptions, the firm’s own rules of thumb that only surface through real interaction with the system…
…A horizontal agent could in principle build the same learning infrastructure. The reason it doesn’t, beyond pure focus, is UX: capturing this kind of knowledge depends entirely on the workflow surfaces you give the user, and vertical players can shape those surfaces around exactly what their workflow needs to surface. Horizontal tools can’t. Eval sets, labeled outputs, and edge-case taxonomies can compound into a vertical-specific data flywheel which can fuel fine-tuning the next entrant can’t generate without comparable production exposure. Whether this is possible depends on data rights, the volume of production exposure accumulated, and the structure of customer contracts, but pattern recognition accrues regardless.
Managing model variability and complexity: The labs are already routing internally — different model classes for different requests, ensembles under the hood. What they can’t do is route across vendors, or evaluate a competitor’s model for a specific sub-task, or use an open-source fine-tune for the narrow piece where it’s actually best. The Rest of Oz company picks the right model for each sub-task across the entire model market, not just what its parent lab ships. It also does the work nobody wants to do — re-running evals on upgrades, recalibrating prompts for the customer’s edge cases, rolling out without breaking production — every time a new model lands. The labs aren’t doing this on the customer’s behalf; they sell you their next model and tell you to migrate…
…Cost optimization: Running every query through Opus 4.7 is the fastest path to negative gross margins. The best Rest of Oz companies route across tiers of models — frontier models for the hardest tasks, mid-tier for the bulk, smaller custom or fine-tuned models where they’ve earned the right to use them. Some are now post-training their own models on top of that, optimizing them for the narrow slice of work their customer cares about and serving them at a fraction of the cost of a frontier API call…
…Governance: There is considerable value in becoming the control plane for how their customers run AI in that vertical – the place where permissions, auditing, what-the-agent-is-allowed-to-do, and what-the-agent-actually-did all converge. That control plane is built out of use case specific guardrails that look completely different across industries and job types. Because they own the tools, the workflows, and the data the agent touches end-to-end, they can provide deterministic outcomes in ways horizontal tools will struggle to. They are also the entity that absorbs the regulatory complexity for the end buyer — FRCP and bar rules in legal, HIPAA in healthcare, SEC and FINRA in finance, state insurance regulations, and so on. A horizontal player can’t credibly do that without becoming a hundred different verticals at once. CIOs want to have a partner that contractually states they are handling compliance for the agents they are providing.
All of these come back to the same thing: focus. That could be a vertical (insurance, legal, accounting) or a function done deeply (sales, customer support, finance). Either way, the work needs a team that’s heads-down on one customer set — its workflows, its edge cases, its regulations. The labs aren’t built for that. They have to be everywhere, for everyone, which is how they built the Yellow Brick Road in the first place.
3. Sergey Brin: Where Frontier AI Is Headed | Unscripted Q&A @ AGI House × Google DeepMind (Transcript here) – Rocky Yu and Sergey Brin
Sergey Brin: That’s a great question—what’s next after we hit AGI? Everybody is pretty focused on accelerating the growth in AI right now. You’re right: we started with the web and internet search, went through the mobile generation, which was another big explosion, and now AI is a huge new industry trend. What comes after that? I think if you can answer that, you’ll have a fantastic company on your hands…
…Audience Member: I have two questions. First, now that we talk about superintelligence, and AI can help us drive cars and do office work—what kind of thing do you think only humans can do after superintelligence? Second, 20 years ago Google was famous for connecting people, and now it’s a company focused on AI. So my question is about strategy: what do you think Google’s role will be over the next 20 years?
Sergey Brin: Small questions, I guess—what is humanity’s role in this world, and what is Google going to do for the next 20 years? The definition of intelligence has always shifted with what machines can do versus what people can do. For a long time, chess was the measure of intelligence, and then Deep Blue beat Kasparov in the 1990s. The interesting thing is that people kept playing chess. How many people here know who the top-ranked human chess player is? Anyone can yell the name—I’m assuming it’s Magnus Carlsen, and people bounce up and down. But how many know the top-ranked AI program?
Audience Member: Stockfish?
Sergey Brin: That’s the most popular—is it number one? You don’t think AlphaZero can beat Stockfish? Okay, well, you’re the only one who named the top chess program; let’s point that out. My point is that computers doing things well hasn’t stopped humans from getting better and better at them, getting more recognition, and enjoying them. We’ve adjusted our view over time—it used to be that chess was the intelligent thing, then Go was the intelligent thing, then poetry or painting. I think we’re going to find that AIs can do a whole lot of surprising things, but they also help advance people in doing those things. Since AlphaGo, the game of Go has advanced a lot—the players who played against Lee Sedol became vastly better afterward, and Ke Jie did too after he played AlphaGo. It pushed the state of the art. So people will be able to enjoy and do a lot of things even with AI assistance. As for the 20-year question—I don’t know. I think we should let somebody else ask. That’s a big one.
Audience Member: Do you believe transformers are sufficient for AGI?
Sergey Brin: Great question. I’ve asked myself that a bunch of times. Transformers have been weirdly flexible—we use them for image and video in addition to text, and they’ve exceeded their original capability. To be fair, they’ve also changed along the way: we have sparse transformers and a lot of little details that have shifted, so it’s not exactly the same thing as the transformer paper. If I had to guess whether something close to that could be AGI, I’d say yes—just because they’ve been able to evolve so much. But they are changing; it’s not the exact same thing as the original transformer paper…
…Audience Member (Boris): What’s your perspective on how world models can help reach AGI?
Sergey Brin: World models are basically video models. People talk about AGI pretty broadly. I think of AGI as the idea that the AI can actually improve itself. Other people—and they’re probably more correct—think AGI means the AI can do anything a person can do. Those are two different things. To do anything a person can do, you absolutely need to understand and interact with the physical world. So being able to dream or imagine what’s going to happen in the world if you do something, and to comprehend it, is obviously important. If you’re going to do everything—and that extends to robotics—world models are key. You all have probably had more time to play with our Gemini Omni model than I have, honestly, because I’m deep into the self-improvement game. But we’ve been working on that for a long time, and Omni is the latest version. Omni is also pretty cool because it’s the same Gemini—we train it with all the text and all the other things, exactly the same way. The fact that these converge is amazing. But yes, you need that capability for the ability to interact physically.
4. Blackstone Investors Ask to Pull $4.4 Billion From Private-Credit Fund – Matt Wirz
Investors in Blackstone’s flagship private-credit fund, known as Bcred, asked to redeem 10% of their shares in the second quarter, up from about 8% in the first quarter. That amounted to investors asking for $4.4 billion.
Blackstone will limit redemptions from the $79 billion fund to 5%, a reversal from its strategy in March when it opted to pay the full amount requested. The about-face highlights rising financial strain on managers of large private-credit funds marketed to individual investors who continue to ask for their money back…
…“BCRED remains well capitalized, and repayments [from loans] and inflows have outpaced shares repurchased,” the firm said Thursday. It said the fund’s structure, allowing it to limit redemptions, is a core feature that is meant to trade some liquidity for long-term performance…
…Wealthy individuals piled into private-credit funds—known as business-development companies, or BDCs—which invest in high-interest loans to midsize companies and distribute most of the income they collect to shareholders via dividends. The boom ended this year when investors turned bearish over increasing loan defaults and the potential for future losses from lending to software companies.
The Blackstone fund is the largest of the bunch, surging to a high of $82 billion at the end of 2025, but it is now shrinking, cutting into the fees the firm can collect.
5. The AI Price War Is Here, Piling Pressure on OpenAI and Anthropic – Bradley Olson and Tina Li
Big companies and startups, chafing at rapidly escalating artificial intelligence costs, are increasingly turning to tools that tap in to cheaper AI models, including some from China. That’s raising pressure on industry leaders OpenAI and Anthropic to lower their prices, a prospect that could hurt their ability to grow into profitable enterprises…
…The ecosystem allows autonomous AI systems, or agents, to use cheap models—including those made by Chinese companies like Alibaba and DeepSeek—for many functions. The agents only tap the most capable versions of OpenAI’s ChatGPT and Anthropic’s Claude for more complex tasks. That can reduce costs for some AI-assisted work by as much as 95%, according to executives using the tools.
“Once we find something that is working well and engineers love, we find ways to make it cost effective,” said Dan Robinson, founder of Detail, a startup that identifies bugs. “There’s really an embarrassment of riches right now coming out of the open source labs.”
Robinson shifted 90% of Detail’s workload from Claude and Google’s Gemini to custom models and GLM, a family of models developed in China…
…OpenAI is considering drastic cuts to the prices it charges AI users, ahead of similar cuts the company expects at Anthropic, The Wall Street Journal reported. The company sees itself as having an advantage in such a scenario because it spent massive sums in the past year to secure access to computing resources at far lower prices than what’s available now…
…Open-source Chinese models have been rising in popularity across American businesses. DeepSeek’s share of AI usage rose from 1% in April to 17% in May on the startup Vercel’s platform, the company said.
On OpenRouter, another startup that processes AI queries, DeepSeek has been the most-used AI company since mid-May. Among their highest-spending customers, open-source token usage grew four times faster than closed-source between fall 2025 and spring 2026, OpenRouter said. The company has also seen more than 500 organizations swap from proprietary to open-source models…
…Anthropic’s recently-released Fable 5 model is more than 50 times more expensive per token than DeepSeek’s V4 Pro, for example.
But the top proprietary models from companies like OpenAI, Anthropic or Google remain four to six months ahead of open-source competitors, researchers say. In some cases that means they can complete a complex task using fewer tokens, equating to a lower total cost…
…Many companies have begun to design their own AI models using open-source alternatives and say they are managing to reduce AI costs. When companies build in-house models and train them with company data, their performance can improve or even exceed the capabilities of frontier AI models, executives say.
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). Holdings are subject to change at any time.