Category: Thinks

Kai-Fu Lee: “In America you have OpenAI, Anthropic, Google, and xAI, each of which believes they’re the genius that will beat everyone else and win the Nobel Prize by solving the ultimate problem of AGI. But the Chinese approach is different. The approach is more like a study group, where one company publishes a model, and the other looks at and plays with it. Maybe even talks to the company about how they trained it. All the members of the study group are building open source and then sharing it. So the study groups are formed of very smart kids who are all funded by companies that still want to show profit every quarter. This is very different from the situation in the US, where companies do not care about returns. In China, companies are constrained in how much they can spend. Alibaba isn’t going to lose $10bn the next quarter. But OpenAI can. So, all these reasons cause the Chinese companies to behave the way they do with modest resources, learning and improving, working as a study group, as opposed to the American winner-take-all strategy.”

NYTimes: “If millions are turning to AI as therapists and friends, it is evidence of a great emptiness many people feel living in modern society. AI is filling needs that mostly shouldn’t exist in the first place if we were a healthy society. People (not talking about government) could willingly spend their energies providing better ways to meet each other’s emotional needs and to fill these voids, yet our energies are focused much more on building and being users of advanced technologies. It’s a backward priority.”

Tyler Cowen: “My core model is both simple and depressing. Fertility rates have declined around the world because birth control technologies became much better and easier to use. And people — women in particular — just do not want that many kids…A lot of women, once they face the realities of the stress and trying to make ends meet, want only one kid. You end up with a large number of one kid families, some people who never marry/procreate at all, and a modest percentage of families with 2-4 kids. There are also plenty of cases cases where the guy leaves, self-destructs, or never marries, after siring a single child with a woman. That gives you the fertility rates we are seeing, albeit with cultural and economic variation.”

WSJ: “Yale University finance professor James Choi recently developed a formula that recommends an asset allocation based in part on your age, income, savings and risk tolerance…In many scenarios, the formula recommends a more aggressive, stock-heavy portfolio than other popular guidelines. It also incorporates more of people’s financial circumstances than common rules of thumb for equity allocation, such as the classic 60/40 division of stocks and bonds or subtracting your age from 100. Many simply outsource the decision to a target-date fund based on the year they plan to retire.”

NYTimes: “For top athletes, grappling with failure is a job requirement, said David Fletcher, a professor of human performance and health at Loughborough University in Britain. A key difference between the best athletes and the rest of us is that they see challenges as opportunities for growth rather than threats, he said. While being naturally optimistic or conscientious helps, researchers say that to some extent, resilience can be learned. What we might perceive as mental Teflon in top athletes is also the result of a lifetime of practice…Just as psychologists have athletes visualize their wins, they also ask them to imagine all the things that could go wrong, and how they’ll respond, said Jessica Bartley, senior director of psychological services for the U.S. Olympic & Paralympic Committee.”

Manu Joseph: “Talking to strangers takes effort and a tolerance for awkwardness. But although small talk can be dull, it could be quite interesting if we take the risk of appearing somewhat unsophisticated. Here are some tips.”

FT: “Protein is an essential part of a balanced diet, helping to grow, strengthen and repair muscle. Until recently, US dietary guidelines recommended about 46 grammes a day for women and 56 for men.  But eating more protein is in vogue thanks to popular diets, the power of social media influencers and the advent of GLP-1 weight-loss drugs, which can lead to muscle loss as people eat smaller portions.”

WSJ: “Some of the most successful restaurant turnarounds have come from winning a new time of day. Fixed costs like rent and utilities don’t change, so boosting traffic during slow hours is far more profitable than cramming more customers into already-busy shifts. McDonald’s famously revolutionized fast-food economics by adding breakfast. This move featured such items as the Egg McMuffin and coffee, which have a much higher margin than beef. Breakfast became so popular that when the company made it an all-day staple, growth exploded. Starbucks is now trying the inverse: turning itself into a credible all-day snack-and-beverage business.”

SaaStr: “The 2026 SaaS crash is real. But it’s not because AI agents are going to replace Salesforce next quarter. It’s because: AI is eating the budget, growth has been declining for three years, [and] the market finally stopped pretending otherwise.”

FT: “For over a decade, economists have grappled with a modern iteration of the Solow Paradox: we have seen artificial intelligence everywhere except in the productivity statistics. Sceptics argue that the reason for this is that modern innovation in machine learning systems and now generative AI pale in comparison to the great inventions of the past. However, the latest benchmark revisions from the Bureau of Labor Statistics suggest the statistical fog may finally be lifting. Data released [recently] offers a striking corrective to the narrative that AI has yet to have an impact on the US economy as a whole. While initial reports suggested a year of steady labour expansion in the US, the new figures reveal that total payroll growth was revised downward by approximately 403,000 jobs. Crucially, this downward revision occurred while real GDP remained robust, including a 3.7 per cent growth rate in the fourth quarter. This decoupling — maintaining high output with significantly lower labour input — is the hallmark of productivity growth.”

Aditya Puri: Our long-term plan was always focused on our customers’ needs, on business demands, and on what our competitors were doing. The issue for us was always clear: We were going to follow our plan. If a short-term opportunity came up, we analyzed the risk. We decided that, at any point, 25 percent of the bank’s business had to come from areas we weren’t already in. For example, once corporate lending was established, 25 percent had to come from SMEs. You’ve got to analyze the risks: Have you priced properly for it? What is the probability of default? What’s the cost-to-revenue ratio? What’s the total portfolio going to be? Those were always in mind. If a short-term opportunity would compromise any of those, I’d let it go. Frankly, most of the time, unless it was very clear-cut, we would pass. Acquisitions were a different animal, but they followed the same principle. When we were considering Times Bank [in 2000], we were accretive from day one—we saw the customers, the branches, the capital they had, and the price made sense to us. The same applied to our other acquisitions. Once you acquire a business, you need to be sure you’re achieving what you set out to do. They had to fit into our systems and follow our culture, including our values of trust, transparency, and customer focus.”

FT: ““The simplest way to think about Claude Code is that it is a chatbot that can do stuff,” said Guillaume Princen, Anthropic’s head of digital native businesses. “What Claude Code was for developers, Cowork is for knowledge workers,” he added.”

NYTimes: “Karl von Randow and Matthew Buchanan created Letterboxd in 2011, but its popularity ballooned during the pandemic. It has grown exponentially ever since: Between 2020 and 2026, it grew to 26 million users from 1.7 million, adding more than nine million users since January 2025 alone…If Rotten Tomatoes has become a tool of Hollywood’s homogenizing marketing machinery, Letterboxd is something else: a cinephilic hive buzzing with authentic enthusiasm and heterogeneous tastes. The platform highlights audiences with appetites more varied than the industry has previously imagined, and helps them find their way to movies that are substantial. Black-and-white classics, foreign masterpieces and forgotten gems are popular darlings, while major studio releases often fail to find their footing. In an online ecosystem dominated by the short, simple and obvious, Letterboxd encourages people to engage with demanding art. Amid grim pronouncements of film-industry doom and the collapse of professional criticism, the rise of Letterboxd suggests that the industry’s crisis may be distinct from the fate of film itself. Even as Hollywood continues to circle the drain, film culture is experiencing a broad resurgence.”

WSJ: “When [Daniel Coyle] talks of flourishing, he talks of vitality and awakening, but also of slowing down and relaxing. He sees value in observing the pauses between things—intervals that, he argues, can be as eloquent and productive as any bursts of energetic activity. He admires good-humored, self-deprecatory attitudes and a loose, collaborative approach to problems. In all of this, he’s getting at something powerful but hard to quantify that we all perhaps have felt, if only briefly, at one time or another.”

Steven Sinofsky: ” AI-enabled or AI-centric software is simply moving up the stack of what a product is. Software did not create online banks. Banking always required software. Software that faced a consumer, banking or traveling or shopping or reading or viewing, just became an essential part of the bank, travel, etc stack. Sometimes this created new from-scratch companies and sometimes it created new companies inside old ones. Industries were restructured as assets moved around. However big and complex you think a legacy business is today, it will be vastly larger and more complex tomorrow, and it will do vastly more. Think I’m crazy? Consider what banking was like in 1995. If you have any experience you know your choices, features, options, etc were one-thousandth of what you have today, even if fundamentally you got a paycheck, paid your bills, and might have had a credit card.”

FT: “In the history of friction-removal, physical toil fell first. Washing machines erased scrubbing while plumbing and thermostats ended the hauling of water and tending of fire. Then came bureaucratic easing: e-filing, direct deposit, money transfers and online flight check-in all swept away the drags that breed inefficiency. Next, distance and delay. GPS made it nearly impossible to get lost and one-click online ordering eclipsed errands. Streaming put an end to trips to the video store and then the store itself. Remember dot-matrix printers? The clatter as each line crawled across the page, leaving behind an embossed texture, the perforated edges you tore off and the ribbon you changed when it faded. Inkjets and lasers made them obsolete. PDFs came next, and then chatbots, which make even reading the PDF optional. Each advance made the tasks simpler, then pointless, then delegable.”

FT: “How Africa Works is arranged in three parts. The first, contrary to the title, is an analysis of why Africa doesn’t work. More accurately, it catalogues the factors, sometimes surprising, that help explain why most of the 54 states into which Africa was corralled by colonialism have failed to emulate Asia’s economic take-off. The second section is a study of four states — Botswana, Mauritius, Ethiopia and Rwanda — that have managed to generate long periods of sustained growth. The third is an assessment of what it would take for other African economies to emulate that record, with particular emphasis on the agricultural and manufacturing revolutions that were essential to Asian growth. Throughout, Studwell steers carefully between the Scylla of fatalism and the Charybdis of frothy optimism.”

Zack Kass, author of The Next Renaissance: AI and the Expansion of Human Potential : “A lot of the job fears are overblown. On a macro basis, they’re terribly overblown. That’s because if we automate most work, something good will happen. I remind people of this all the time. Every time we put a small farmer out of business, it means that far more people could be fed by industrial farming. One could bemoan the loss of something yet acknowledge the economic gain that lifts people out of poverty. If one wants to go back to small farming, one can. It just means that it will come at the cost of hundreds of millions of lives. If you return to that hyperlocal farming, a lot of the world will plunge back into food scarcity. Now there is a panacea. You could have organic living, local living, but we aren’t there yet technologically. That probably requires fusion, desalinization, small modular reactors, and more. I would argue that the problem we are presented with is that the uncertainty in the market alone is casting an enormous shadow. I challenge people to see it as an opportunity.”

Elon Musk: “The availability of energy is the issue. If you look at electrical output outside of China, everywhere outside of China, it’s more or less flat. It’s maybe a slight increase, but pretty close flat. China has a rapid increase in electrical output. But if you’re putting data centers anywhere except China, where are you going to get your electricity? Especially as you scale. The output of chips is growing pretty much exponentially, but the output of electricity is flat. So how are you going to turn the chips on? Magical power sources? Magical electricity fairies?…It’s harder to scale on the ground than it is to scale in space. You’re also going to get about five times the effectiveness of solar panels in space versus the ground, and you don’t need batteries. I almost wore my other shirt, which says, “it’s always sunny in space”. Which it is because you don’t have a day-night cycle, seasonality, clouds, or an atmosphere in space. The atmosphere alone results in about a 30% loss of energy. So any given solar panel can do about five times more power in space than on the ground. You also avoid the cost of having batteries to carry you through the night. It’s actually much cheaper to do in space. My prediction is that it will be by far the cheapest place to put AI. It will be space in 36 months or less. Maybe 30 months.”

FT: “Knowledge loss is counterintuitive because we live in a world overflowing with learning. That pushes us to think of knowledge as infrastructure: a building that, once erected, will stand for ages. But knowledge is both alive and fragile. It is embodied in people, and the teams and communities they form, and is transmitted through repeated, almost ritualistic practices. Imagine trying to train a new generation of surgeons in a world where no surgeons remain. While that is a particularly dire example, it is reasonable to say that knowledge is like a shark: it must keep moving to stay alive.”

FT: “For many people, the most meaningful careers do not follow straight lines. I spent two years as an engineer and three years on Wall Street before I found my place in venture capital. At the time, those moves looked like detours. In hindsight, they were formative. They provided perspective, skills and clarity that no carefully crafted résumé ever could. That experience is not unusual. In my research I have found countless examples of successful people who switched careers midstream. It often takes time to figure out what work truly makes you happy. When you are driven by curiosity — as opposed to society’s conveyor belt — your career feels less like work. When you love what you do, the more you want to do it and the better you will become, increasing your chances of success, financial and otherwise. This is where the conversation needs to shift, especially for parents. Wanting your child to be financially secure is understandable. Even the great Willie Nelson suggested mamas make their kids “be doctors and lawyers and such.” But in today’s world, that instinct can backfire.”

WSJ: “Today, Nvidia is nearly 20 times as valuable and five times as profitable as IBM was back then, adjusted for inflation. Yet it employs roughly a 10th as many people. That simple comparison says something profound about today’s economy: Its rewards are going disproportionately toward capital instead of labor. Profits have soared since the pandemic, and the market value attached to those profits even more. The result: Capital, which includes businesses, shareholders and superstar employees, is triumphant, while the average worker ekes out marginal gains. The divergence between capital and labor helps explain the disconnect between a buoyant economy and pessimistic households. It will also play an outsize role in where the economy goes from here.”

Shankkar Aiyar: “India is an outlier among outliers. It has vaulted from the 10^th largest economy to the fourth largest in the world in just over a decade. It has evaded the law of necessary and sufficient conditions. It is the only economy to reach the status of the ‘fourth largest economy’ without the power of global brands and banks of global size to fund market expansion.”

Mark Roberge: “The current AI-native B2B software startups face significant risk from foundational model expansion, circular revenue redundancy, high valuation overhang, and first-mover (dis)advantage. A potential bright spot in the cohort may be vertical AI startups, where lower TAMs and higher regulatory compliance product requirements shield them from foundational model disruption, valuation overhang, and CIO “build” tendencies. In the long term, early followers starting in the “trough of disillusionment” phase may take their respective categories at the expense of the first movers.”

Z Reitano: Economics of a Super Bowl ad.

SaaStr: “Don’t try to compete on models. Compete on what you build on top of them. The winners will be the companies that deeply understand specific workflows in specific industries and build AI that solves those problems end-to-end.”

FT on life in a depopulating world: “Picture your street in 2100, presuming climate change hasn’t washed it away. With half the population, some homes have been combined into vast single residences. Other buildings have been torn down, to make way for cooling mini-jungles. The future inhabitant of your home, born in 2026, now in their seventies — which by that time might be considered middle age — could find life cornucopian.”

Siddhant Khare: “If you’re an engineer who uses AI daily – for design reviews, code generation, debugging, documentation, architecture decisions – and you’ve noticed that you’re somehow more tired than before AI existed, this post is for you. You’re not imagining it. You’re not weak. You’re experiencing something real that the industry is aggressively pretending doesn’t exist. And if someone who builds agent infrastructure full-time can burn out on AI, it can happen to anyone.”

Suzanne Vranica: “When you set out to do an ad, you want to brief all the constituents. You want to put out a brief that says, “This is what I’m looking for.” You’ve got AI doing all of that grunt work right now. You’ve got AI doing production work. We used to have storyboards. Now you can get AI to basically create an ad. It’s not going to run on air, but it’s something that you can present to a client or you can get a feel for and you can tweak it and then really film it. So it’s that underbelly that it’s playing right now, but this technology is improving so quickly. There are companies like Mondelēz and others that have spent the last two years feeding these bots, all of their old advertising, all of their brand books and positioning so that these things get smarter. And those are the companies when they decide to turn it on and say, “Go create an ad,” you’re not going to be able to tell the difference between what’s now called AI slop and what looks like a perfectly created spot. We’re a little bit away from that at this point, but we’re going to get there.”

NYTimes: “Sonja Lyubomirsky has been a leading researcher on the science of happiness for decades. And for just as long, people have asked her: What is the secret? Dr. Lyubomirsky, a distinguished professor of psychology at the University of California Riverside, has always chafed at this question. The secret to happiness? How ridiculous and reductive. When pressed, she told me she tends to say something along the lines of: “Connection and relationships. Positive thinking, which includes gratitude. And a sense of control in your life.” But if she really had to choose one thing, she said, the secret to happiness is “feeling loved.” That’s the premise of her latest book, “How to Feel Loved,” out today, which she co-wrote with Harry Reis, a professor of psychology at the University of Rochester who studies close relationships.”

WSJ: “Cellphones and online sports betting were made for each other, because you now have a casino in your pocket. The problem is that casinos always stack the odds in their favor. This built-in profit is the vig, or vigorish. Bet a dollar on an even game and you can only win 90 cents. In props and parlay bets, the vig might be 25% or higher. Prop bets are made on a team’s or player’s individual statistics in a game, like whether LeBron James will score more than 25 points, while parlays combine multiple bets—that four teams will all win their games, for instance—into one bet. Both are attractive. Casinos appreciate the higher vig and gamblers love the prospect of a big payout on a small bet. Never mind that the odds amount to a tax on people who can’t do math.”

FT: “The rise of software in private equity can be traced to two once little-known specialist buyout firms, Vista Equity Partners and Thoma Bravo. The leaders of both firms emerged as prolific acquirers in the wake of the 2000s dotcom bust as they acquired mid-sized software companies selling cyber security, or services to niche industries such as hospital systems, car dealerships or parking meter networks. Many of their targets had been abandoned by public investors but came with steady growth and reliable customers who rarely pushed back on annual price increases. Vista and Thoma earned large gains from their early funds, generally in excess of three times investors’ initial investment after fees. Within the PE industry, the two firms’ returns trounced those of larger, established funds that participated in a mid-2000s bubble of mega-sized takeovers of companies that quickly collapsed.”

TheMaxSource: “The International Software Benchmarking Standards Group analyzed thousands of software projects and found that teams of nine or more are significantly less productive than smaller teams. This isn’t a small effect. The data shows a clear inflection point. Below nine people, productivity holds steady or improves. Above nine, it drops. Analysis of software projects shows teams of five to seven people had the highest productivity with approximately nine percent less variation than teams of three to five, and 12 percent less variation than teams of 1.5 to three people. The five-to-seven range delivered the best schedule performance and lowest development effort. Larger teams translated directly into higher costs and longer timelines.”

Arnold Kling: “My sense is that the near future belongs to the sorts of people who thrive at Alpha. Having the ability to work with AI and the ambition that comes from internal pressure will enable some people to accomplish far more than they could have previously. Those who resist AI and/or fail to set high expectations for themselves will have to settle for lower income and status. The next several years are going to find us living out “average is over.””

TheMaxSource: “McKinsey surveyed 1,200 business leaders and found that 61 percent say at least half the time they spend making decisions is ineffective. Fewer than half report that decisions are timely. The problem isn’t lack of information or bad analysis. The problem is waiting for everyone to agree. Organizations treat consensus as the gold standard for decision quality. Get everyone in the room. Hash it out. Make sure nobody objects. This sounds reasonable until you measure what it costs. Companies that optimize for agreement optimize for delay. Delay means competitors move first. Markets shift. Opportunities close.”

David Brooks’ book recommendation: ““Man’s Search for Meaning” by Viktor Frankl. He teaches that the animating human drive is not the desire for money, fame or popularity. It is the desire to feel that your life has purpose.”

Vasant Dhar: “In the new era of modern AI, I am forced to consider whether I am becoming obsolete, both as a practitioner and professor. Many of us should be asking ourselves the same question: how do I stay ahead of the machine?”

FT: “Digital advertising now accounts for about three-quarters of the more than $1tn global ad market and Google is the dominant player in the sector, with advertising revenues of $82.3bn in the fourth quarter of 2025, up about $10bn from the previous year. However, having siphoned off revenue from traditional TV and print advertising markets, Google and other giants such as Meta and Amazon are now facing threats to their well-oiled advertising machines. Google has already introduced advertising into its AI mode, a chat interface that sits within traditional search, and AI overviews, its summaries posted at the top of search results.”

SaaStr: “Here’s the counterintuitive insight that’s reshaping how the smartest AI founders think about unit economics: Your inference costs aren’t your gross margin problem. They’re your CAC replacement.  If … You Have a Product That Can Truly Deliver with AI. The companies growing fastest right now—Cursor crossing $1B ARR with ~300 employees and no traditional marketing, Lovable and Replit both hitting ~$300M ARR with zero paid acquisition, Harvey and Legora in legal, OpenEvidence in healthcare—aren’t sweating inference costs. They’re optimizing them for sure, but they are also leaning into them. They’re treating compute as their primary growth investment, not their primary margin drag.”

Andrej Karpathy: “It is hard to communicate how much programming has changed due to AI in the last 2 months: not gradually and over time in the “progress as usual” way, but specifically this last December. There are a number of asterisks but imo coding agents basically didn’t work before December and basically work since – the models have significantly higher quality, long-term coherence and tenacity and they can power through large and long tasks, well past enough that it is extremely disruptive to the default programming workflow.”

FT: “The internet made information global. Crypto is making a similar impact on money. While recent headlines might fixate on prices of bitcoin, a deeper and longer-lasting shift is under way in digital payments. This is the year that stablecoins, or cryptocurrencies pegged to assets such as the dollar, are becoming part of the mainstream for online and international payments. Call it money’s “WhatsApp moment”. Just as chat apps like WhatsApp collapsed the cost of international messaging from, say, 30 cents per text to zero, stablecoins are doing the same in financial transactions. The numbers bear this out: stablecoins moved over $12tn in value last year, after filtering out bots and other inorganic activity — volumes that are rising towards Visa’s $17tn of transactions last year but made at a fraction of the cost. In the process, stablecoins are bringing the internet’s original vision of openness and interoperability to finance. Given how blockchain technology allows stablecoins to be programmed, money is in effect becoming software.”

WSJ: “In past eras, the perfect weapon was one that would inflict maximum damage on an enemy. A bigger cannon, a more explosive shell, a better strategic bomber—a B-52 instead of a B-29—or a hydrogen bomb instead of an atomic one. In “The Warhead” Jeffrey Stern…traces the evolution of precision-guided munitions (PGMs) over the past half-century. By Mr. Stern’s account, the PGM represents a technological breakthrough that has changed history and the way wars are fought. It’s an absorbing tale of scientific and technological achievement, political and military hubris, and the story of ordinary people being caught up in other people’s bad decisions—and dying as a result.”