Category: Thinks

FT: “Scientists have unveiled synthetic cells aimed at revolutionising industrial production by serving as tiny biological factories to make environmentally friendly materials for goods from drugs to plastics. The new invention known as SpudCell will be made available to researchers around the world, as part of an expanding effort to put micro-scale bioengineering at the heart of lower-carbon manufacturing for everyday items. SpudCell is the first synthetic system to achieve a complete cell cycle of feeding, growth and reproduction, according to a paper released on Wednesday but not yet peer-reviewed. It is built from the bottom up using known non-living chemical components, as opposed to previous methods based on pared-down living cells and their naturally occurring parts.”

NYTimes: “The discovery of electricity did not just beget lightbulbs; in time, it enabled the modern mass production system and the entire vast digital revolution. A.I.’s transformations may be even more sweeping. But generative A.I. as it currently exists cannot easily replace human beings, because it cannot manifest human intelligence. That won’t stop it, however, from destabilizing society in ways more profound than we might even imagine. The sooner we update the way we think about the current state of A.I., the sooner we can all stop freaking out about the wrong things — and start preparing ourselves for the ways it really will transform our world.”

Ming-Chi Kuo: “The memory supply-demand gap will keep widening through 2027.”

FT: “The difference lies not in the instincts of the strongmen but in the robustness of the systems they operate in. If a strongman ruler is already governing in a largely authoritarian system…they are almost impossible to get rid of. It might take a coup or an unusually effective ruling-class conspiracy to make a change at the top. But if the strongman has come to power in a system that still has free elections — as well as courts that retain some independence and a military that will not follow unconstitutional orders — then it is still possible to evict them from office.”

FT: “Because America is based on an idea, rather than ancestral ties, its revolution triggered a quest that can never be exhausted. Much is rightly made of the fact that many of the founders, notably America’s first and third presidents, George Washington and Thomas Jefferson, owned slaves while proclaiming the equality of men…The US example helped inspire freedom movements around the world, including the Spanish American wars of the early 19th century that liberated much of the hemisphere from colonial rule. The same creed helped insulate the US from the lure of foreign ideologies, notably Marxism and fascism, and motivated its prosecution of the cold war. It is no coincidence that a nation founded on ideas became the world’s leading source of new ones, commercial and political.”

Bruce Feiler: “Everyone is leaving jobs, losing relationships, researching strange diseases, and feeling like we’ve abdicated our relationships for our smartphones, in a lot of ways. Everyone is experiencing a type of existential home sickness. That’s when I realized that I wanted a way to reconnect, to remake all of these things and restitch the bonds in my life. I thought, “I need a ritual.” Rituals are one of the few things that we know work to hold us together. A ritual is a shared unnecessary act—shared because it connects with other people—that makes us feel at home. They are unnecessary acts because you don’t need to get down on one knee to propose or wear black to mourn. Yet, they become necessary to give us meaning. They also make us feel at home.”

NYTimes: “Private credit is essentially a rebranded version of high-interest-rate lending that was first popularized under the term “junk bonds” in the 1980s and later became a subset of “distressed” or “special situations” investing. In prior iterations, these practices centered on banks, which amalgamated money from deposits and divvied out loans. This had advantages — it was heavily regulated and relatively transparent — and disadvantages, in that it put the banks’ ordinary savers at risk. Laws enacted after the 2008-9 financial crisis made it far more difficult for banks to be directly involved in such loans. The demand for financing from fledgling companies didn’t disappear, so private equity firms stepped into the void under the name private credit and began lending money.”

Business Standard: “Our monster Consumer India just got more monstrous, with another shot in the arm. After the first phase of liberalisation in the 1990s — when the prices came down and quality went up (due to lower taxes, more competition) — came Chinese goods with great price-performance points and innovative everyday products that had not been seen before. Then came Amazon, with all its incredible merchandise assortment, prices and the ability to return anything you did not want. Now comes quick commerce. Even if QC is eventually contained, it will not be for a lack of customer enthusiasm but for a lack of supplier staying power and patience through the scaling period, or their shifting to a high-margin and high-ticket-size merchandising mix. Consumer India leapfrogs in the most interesting ways, skipping stages of evolution other markets went through. Offer it customer-perceived superior value and do not demand that it change its ways, and you have a hit.”

Paul Graham: “One of the biggest advantages of AI will be that it lets companies get further before they cross the lines (at about 10 and about 150 people) beyond which groups become less productive.” [via Arnold Kling] More from Arnold: “Experienced software developers are used to working in teams. They have gotten accustomed to relying on Cursor. But at some point, we will see the emergence of software developers who are used to working alone. Cursor appeals to software engineers who are used to working without AI and who want to be able to see the code. My prediction is that in a few years the best software engineers will trust the best coding AI’s Instead of Cursor, what will ultimately emerge are tools that optimize for communication between a lone developer and AI.”

FT: “Uniqlo releases only 800 new designs per year, and only about half of those are changed in its six-monthly renewal cycles. By contrast Zara, a fast-fashion pioneer, churns out thousands of designs every year, with new iterations arriving every fortnight. At Shein, thousands are released each day. Smaller ranges mean Uniqlo might end up selling a million units of a single T-shirt design versus, say, less than 100,000 at fast-fashion houses, generating scale and cost advantages when it comes to raw materials, sewing and dyeing. The resulting high-quality products at reasonable prices were a winning model in the deflationary market of post-bubble Japan. Some analysts believe its lower gross profit margin relative to Inditex indicated how Uniqlo passed the price and quality benefit on to its cost-conscious consumers.”

NYTimes: “[A study] suggests that the most dangerous times for investors are when the market is high — and we may be in such a time right now. The study, by Hendrik Bessembinder, a finance professor at Arizona State University, shows that most of the biggest losers since 1926 were tech companies. They included stocks that boomed during the dot-com era and in the halcyon days just before the financial crisis that began in 2007 — and many crashed when those boom cycles ended.”

WSJ: “Unlike in previous tech cycles, corporate adoption of AI rests on all employees—not just developers—picking up on the technology. AI is increasingly being billed by usage, and the price of tokens, the basic unit of AI computing, has been volatile. That all translates to higher costs for AI.”

[Via Arnold Kling]: “The revolutions are led by the educated-but-blocked: young people with enough knowledge to understand the system and enough frustration to want to tear it down.”

FT: “The wisdom of crowds has never lain in the consensus. It lies in the disagreement. Once everyone thinks alike, crowds are error-prone. ”

NYTimes: “Bending Spoons is not the only firm chasing tech’s castoffs. Constellation Software, a Canadian company with $11.6 billion in annual revenue, specializes in buying business-oriented software and technology companies. A Los Angeles company called MediaLab, run by Michael Heyward, the founder of the anonymous social media app Whisper, has bought the assets of Imgur, the image-sharing site; Kik, the messaging app; and Genius, the music lyrics site. Since it was founded in 2013, Bending Spoons has purchased more than 50 companies. Last year it generated $1.3 billion in revenue.”

Ethan Mollick: “Being on an exponential means each change over a fixed window is larger than the one before it. If your organization wrote an AI plan any time before the winter of 2025, it described a system that could do a couple of hours of work with a fairly high error rate. A few months later, you can get sixteen hours or more of work from a single prompt. This is why AI keeps feeling like it is making leaps, even though it is a curve on a graph, we keep experiencing a steady doubling of capability as a series of shocks. We are very bad at feeling exponentials from the inside, and we are currently inside one.”

NYTimes: ‘A slew of factors goes into investments and status in Silicon Valley. Lately, an intangible one has pulled ahead as a predictor of success or failure: “signal.” If your behaviors are deemed high-signal: Congratulations, you’re coming across as a winner and rewards may follow. If your actions are seen as anti-signal: Sorry, you’re radiating cringey energy that may hamper your chances.”

Anindya Chatterjee: “What are engineering jobs that AI cannot do? Jobs that require skill with both computers and hands, travel to jobsites, human presence, unexpected connections between ideas, and situations where data is unavailable or confidential. These, more than entrance-exam ranks and offer-withdrawing recruiters, need thinking about.”

Noah Smith: “Over the last two years, I’ve felt like my job has become a bit less important than it used to be, for three reasons: (1) The rise of populism on all sides of the political spectrum in the U.S. means that smart ideas are simply not as likely to be implemented by the people in power. (2) The general shift to Substack and other monetizable direct-to-audience channels has made punditry less conversational. (3) The rapid proliferation of AI writing has increased the demands on readers’ attention (including my own).”

WSJ: “The memory armageddon has arrived…Buyers of consumer electronics are getting hit. The primary driver is the skyrocketing cost of memory and storage chips, especially those known as DRAM and NAND flash memory, which are essential for transferring data and storing information on devices. These chips are the same ones craved by artificial-intelligence companies, which use them to help train and run large language models, coding agents and other tools. As AI adoption has exploded, the memory-chip industry—dominated by just three companies: South Korea’s SK Hynix and Samsung Electronics and Boise, Idaho-based Micron Technology—is suffering from a major capacity crunch.”

WSJ: “When it comes to nutrition, science is converging on the following recommendations for longevity and health: a diet rich in vegetables, whole grains, nuts and plant-based unsaturated fats, moderate fruit and fish consumption, low red- and processed-meat consumption, and very low ultraprocessed foods and added sugars. While protein consumption gets a lot of attention, research suggests a low but sufficient daily intake of 0.37 gram per pound of body weight, or about 60 grams for a 150-pound person, of which at least 50% is plant-derived, is ideal.”

Paul Kedrosky: “Job loss in the United States is more threatening than anywhere else in the wealthy world. It turns what should be a setback into a potential cascade — income, insurance, mortgage and child care, all at risk at once. Meanwhile, A.I. chief executives won’t stop telling Americans A.I. is coming for them. The technology is a missile aimed at the most fragile part of the American socioeconomic bargain. No wonder Americans are pessimistic about A.I. While better messaging will not fix this, decoupling health care from employment might. Building an unemployment insurance system that replaces income at a meaningful level might. Americans’ pessimism about A.I. is largely rational, about a technology tailor-made to crack their crumbling and antiquated social compact.”

FT: “[China’s] working-age population aged 15 to 64, which peaked at 1bn in the last decade, is due to fall to just 300mn by 2100, according to UN figures — a decline that could prevent China from becoming the world’s biggest economy. Beijing now sees AI-enabled machines as a way out of the demographic trap. Last year the country installed more industrial robots than the rest of the world put together; it also makes most of the world’s humanoids.”

WSJ: “Three questions confront American capitalism at this crossroads, the resolution of which will shape the lives of future generations. Will America deliver economic opportunity to more of its people—or will the gap between the extraordinarily successful and struggling widen? Will America continue building walls to the world—or will it build new bridges? And will America strike the right balance between competition’s creative destruction and the government’s regulatory guard rails—or will it tilt too far in one direction or the other?”

WSJ: “[Consulting] firms have primarily focused on two alternatives to the hourly model: fixed fee and outcome-based pricing. Under fixed fee, firms guarantee a specific, predictable cost for a defined output or project scope, regardless of the hours spent. Outcome-based pricing generally means consultants get paid if they achieve certain mutually agreed-upon metrics for the client, or a range of outcomes based on over- or underperformance…“The shift towards fixed-fee and value-based billing has put considerable pressure on consultancies to produce more output,” said GPTZero CEO Edward Tian. “But if that comes at the expense of basic fact-checking, the reputational damage for the firms themselves and the clients who commission their work is enormous,” Tian said.”

Business Standard: “AI agents are expected to become a new workforce layer, forcing companies to rethink the traditional fresher hiring pyramid that powered India’s IT and consulting industries for three decades. Many executives have been talking about the pyramid reshaping into a diamond, requiring fewer people at the bottom and an AI-native experienced layer in the middle.”

FT: “Holmes is regularly high on the lists of fictional characters most portrayed on screen, up there with Father Christmas and Dracula. Even if you’ve never read one of the stories, you know who he is. And you’re only going to see more of him, because Conan Doyle’s works are now out of copyright, so the character is anyone’s to play with.”

Mint: “India stands at the threshold of a new investment-led growth phase that could redefine its economic trajectory over the coming decade. Much like the investment boom of 2004-08, the country is again seeing the early stages of a broad-based capital expenditure upcycle. This time, however, the foundations appear stronger, more diversified and strategically aligned with long-term national priorities. At the heart of this emerging cycle is a shift towards domestic capacity building. Policymakers are increasingly focused on reducing dependence on imports in critical sectors such as energy, defence, technology and industrial supply chains. Combined with strong domestic demand, supportive reforms and healthy corporate balance sheets, this strategy is laying the groundwork for a sustained rise in investment and economic growth.”

WSJ: “Mr. [Harvey] Mansfield sees in America a “paradox”: The U.S. is both a young, brash nation and the world’s oldest continuous democracy. “We’re a country that needs to renew itself every four years, or find out who should run it,” he says. “It has, within itself, a youth and a maturity. We change our president, which means the entire government, more or less.” America is “an exemplar of something important, which is a successful republic. But it needs to be continually refreshed, as Machiavelli said all regimes need to be.” Mr. Mansfield says America is “exceptional because it accepts the truth of that claim.” We always itch to “go back to our beginnings and see what it was that first made us excited and fresh, and made us seek to be an example to the world.” That can become “a form of bullying,” in which Americans “demand that other peoples do what we advise. And what we advise may not necessarily be for their good, but mainly for ours.” By the same token, “great claims are imputed to us by others, and our failure to live up to them is taken as corruption.””

NYTimes: “Z.ai is on the cutting edge of a wave of powerful but inexpensive A.I. from China that is challenging the lock that OpenAI, Anthropic and Google have had on the industry. Six of the models now on the A.I. leaderboard were developed in China. Z.ai’s new model, GLM-5.2, arrived just as U.S. businesses realized that they had to find ways to cut down on how much they were spending on A.I. It also landed when executives in Silicon Valley were becoming worried that the Trump administration was leaning toward regulating the technology.”

FT: “America’s attractions run deep. It has the best pool of entrepreneurs and US businesses account for just over half of the world’s value creation, or profits in excess of the cost of capital. America also has what Scott Bessent, its Treasury secretary, called this week, the world’s “deepest, most dynamic markets”. Since 2009, its banks, asset managers and traders have grown more dominant globally, and funnel savings into its markets. Citadel Securities, which executes a third of share trading in the US market, says activity is “astronomical”. Larry Fink, the boss of BlackRock, the largest asset manager, had called US capital markets a “juggernaut”.”

The Generalist: “Not every founder was ostracized quite as totally as [Henry] Luce. Culture, class, geography, and even his manner of speaking conspired to separate him from his peers. But the experience of being an outsider, of being profoundly out of place, is the most common trait observed from studying the formations of great entrepreneurs. It is also among the most acute, contributing to a desire to prove one’s relative worth and breeding a sense of specialness, even if reflected through a dark mirror. Perhaps every adolescent feels some moment of difference; this is the beginning of selfhood. But not everyone experiences it quite like this.”

WSJ: “Multitasking is a reality of modern life, yet scientists have long debated what is actually happening inside our brains. For years, they thought the brain couldn’t truly do two things simultaneously. Instead, they believed multitasking was just our brain rapidly switching between tasks. Now, research suggests we really can do multiple things at once.  After enough practice and experience, the brain can rewire its circuitry to support multitasking, according to a recent study in the Journal of Cognitive Neuroscience. The brain can learn to automate certain tasks, by offloading them from an area called the prefrontal cortex—a control center responsible for high-level thinking and decision-making that can only focus on one thing at a time—to other regions.”

Financial Post: “Aside from newcomers, engineering stalwarts Siemens, Schneider Electric SE, Dassault Systèmes SE and ABB Ltd. are embedding AI in their software and automation products to help factories become more productive, more efficient and more competitive. That’s a powerful sales pitch at a time when Europe’s industrial base is under pressure from high production costs, dwindling numbers of skilled workers and growing competition from abroad. Europe’s not alone in trying to seize on the potential. While US firms such as Emerson Electric Co., Rockwell Automation Inc. and Honeywell International Inc. are moving into machine-learning technologies, the biggest challenge comes from China, according to Nicole Lemke, who authored the Interface report. China has been doing “quite well in all things manufacturing in general, but also is very strongly focusing on the intersection of robotics and AI,” she said.”

FT: “What if ideas aren’t getting harder to find, after all?  Large companies may have just become better at hiding the fruits of their labour.”

Azeem Azhar: “The generative AI economy has generated $110 billion in sales over the past 12 months. It is growing fast. On an annualized basis, the revenue run rate exceeds $175 billion.”

NYTimes: “Wordle is a simple game in many respects. The vast majority of users play using the default settings, where you can guess any letter on any turn. But a smaller slice of players (you may be one) choose to play in “hard mode,” which adds a major constraint: Any revealed hints must be used in subsequent guesses. If letters turn yellow and green, you are forced to use them in your next guess. So you might think that hard mode is harder — it says so in the name…An analysis of 730 million games from the last year says the opposite: Players in hard mode solve in fewer turns on average.”

WSJ: “Science suggests you don’t need a long vacation to recharge. A 24-hour break that interrupts your routine—and puts you into a flow state—can be hugely restorative.”

Arnold Kling: “It could turn out that today’s tech stock darlings are also like chain letters. They transfer wealth from the broader public to the initial owners. When the supply of suckers runs out, the share prices will collapse. Even the early owners will not do well unless they cash out in time. The alternative view is that stock prices are high because the companies are creating real wealth. That could be true. It is impossible to prove it false. The very term “real wealth” is undefined. So much of the economy is intangible. We have more stuff, but mostly we are better off because of innovation, specialization, and trade. We are telling ourselves that we are wealthy. The government has borrowed money from us, and we expect to get paid. We own shares of stock, and we think that we can always sell them for a good price. It will all turn out fine in the end. Unless it doesn’t.”

Acquired (via WSJ): “The medium of animation itself is a structural reason why other film studios haven’t achieved Disney’s success: it is uniquely well-suited to form the core of such a flywheel. Mickey Mouse doesn’t age, doesn’t have bad hair days, and never asks for a raise on sequels. Animated characters happily lend the company their name, image and likeness (forever) at zero cost for commercial exploitation. They’re always available to show up at theme parks or on set, and have no problem being in two places at once. Of course, the voice actors and theme park “cast members,” as Disney calls the employees who play the characters, do share in the economics. But it’s quite a different picture than say, Mark Hamill as Luke Skywalker.”