wpboon

  • AI Week

    ‘One day your company announces: stop everything, it’s AI Week. Build something with AI. ‘

    Han Lee, The AI Great Leap Forward

    Obviously, this quote was included my official AI Week introduction slides.

    To sketch the rapid developments in AI since the take-off with Opus 4.5 in November 2025, I showed a bunch of different quotes I’d come across via Hacker News in recent weeks. Some elated, some depressed, some judgmental, some histrionic.

    This one got some laughs.


    So yes, we organized an AI Week.

    I’m not a developer but I lead a team of developers. I’ve been using Claude Code since December 2025, building little apps, automating things, generally feeling the future reaching out from the clouds and putting working software in my hands.

    But nobody on my team seemed to be doing that. And I kept thinking: Why is nobody talking about this? Am I the only one seeing this?

    Make some room

    The answer was obvious, once I stopped to think about it. People might be trying it out on their own time. But during work hours, we ask our people to work on specific features and client requests. There’s no room in that schedule for “go explore new technology.” And you can’t really expect everybody to study this stuff at home.

    At the same time, I kept reading about how AI is going to change software development dramatically. The productivity gains I was seeing myself. I was vibecoding really decent apps in 4-5 evenings. If programming gets that much faster, everything downstream has to keep up. Testing, releases, client contact. The whole pipeline. How?

    Our parent company is also doubling down on AI adoption, which gave me a nice tailwind. But even without that: our team needed to thoroughly engage with the most important technology development of our lifetimes. You can’t watch this one from the sidelines, and you can’t do it in a day.

    So we organized a full AI Week. It was great.

    What happened

    So many chores are baked into software development that we’ve just accepted as the cost of doing business. Things we never got around to automating because we were always too busy building the next thing. Agents can handle a lot of that now. You just have to set them up.

    So after some inspiring demos from advanced users, teams split up and spent two days on self-chosen projects. Real client work continued, it wasn’t a total shutdown. But the focus was on exploring what was possible.

    Non-developer roles surprised me all week. They came in without preconceptions about what “proper code” looks like, or worries about maintainability and testability. They just went straight for what the technology could actually do for them. One customer support officer built an AI toolbox for checking clients’ bug reports, documentation (‘could be a feature!’), tone of voice, etc.

    Beginner’s mind, in the best possible sense.

    Obviously there were also skeptics going in. Some came around. Some still have doubts. That’s fine. The goal was never to convert everyone. It was to experiment as a team and build some shared experience. We did that. And we made a really good start on automating away those chores.

    Once every few months, at least

    I want to do this regularly. Every three to six months. People need time to adapt to new technology — and honestly, even before AI, I hadn’t done enough to just step back and let the team improve their own tools and processes. An AI week creates the permission to do that.

    If you’re thinking about organizing something like this: do it. You don’t need an agenda. You need a week, a shared focus, and people who are curious enough to try.

  • ‘Why I use a Dumbphone in 2025 (and Why You Should Too)’

    ‘While everyone carries a phone in their pocket with a thousand reasons to get distracted, I use a phone without apps, without social media, without distractions. Many would say I’m limited, but I say I’m free.’

    From https://samueleamato.xyz/2025/06/04/why-i-use-a-dumbphone.html

    I keep thinking about ditching my smartphone, but I never do.

    Our’s not a healthy relationship, my iPhone 14’s and mine. It helps to chuck some apps, but I keep finding new ways to distract myself. Currently it’s Dutch news and Hacker News (where I found this, in June of last year, let’s be honest), before that it was Imgur and Vinted, and before that it was Reddit. I’ve owned e-ink devices but they kept dying on me, making them very expensive not-quite tablets.

    The promise of a dumbphone is having more quiet time and being in the moment – but the moment is often awful, boring, or both. Books often can’t hold my attention anymore, and friends don’t come over as often as they did when I carried a dumbphone in the late 90s and early 2000s.

    Anyway, it seems to be working for this guy.

    You have to wonder how that site of his renders on his dumbphone though.

  • Why should Europe have a Tesla?, pt. II

    In February, in response to Pieter Garicano’s ‘Why Europe doesn’t have a Tesla’, I wrote:

    Most Europeans will still prefer to live in the EU, even if growth is slower for the next 10 years, if it means being able to have better living conditions, and less fallout from having to sustain a billionaire class.

    Today, per Euronews: A new way to measure poverty shows the US is falling behind Europe

    As of 2025, the time needed to earn $1 is 63 minutes in the US. This is about twice the average across Germany, France and the UK.

    In Germany, Europe’s largest economy, it takes 26 minutes. In France, the figure is 31 minutes, while in the UK it rises slightly to 34 minutes.

    These figures suggest that average poverty in the US is about twice that of these three countries.

    (…)

    “How can a rich country’s economy grow and yet become poorer?” [Associate Professor of Economics at the University of Oxford Oliver] Sterck asks, referring to the US in his article for The Conversation.

    His answer is simple: inequality.

    I ask again: Why should Europe have a Tesla?

  • Things to read (week 8-2026)

    “It is hardly an exaggeration to say that “the poor” did not exist in classical antiquity.”

    Matthew Walther, Technological Poverty, The Lamp Magazine


    “(We) provide the first causal evidence on how AI adoption affects productivity and employment across more than 12,000 European firms.”

    Inaki Aldasoro et. al., How AI is affecting productivity and jobs in Europe, CEPR.org


    “The first voice of the police report belongs to the nightclub’s doorman, who, through the peephole, observed my mother coming home in her gray coat. They spoke briefly; she told him she appreciated his keeping an eye on things.”

    Louise Bokkenheuser, My Mother’s Body, Virgina Quarterly Review


    “The Barclays, the Lloyds, the Cadburys, the Rowntrees, the Clarks, and the Wedgwoods were all prominent Quaker merchant families. A religious minority that at its peak numbered almost 60,000 people in the country of 6 million – just under 1% – at that time produced an overwhelming share of England’s commercial and industrial infrastructure, so disproportionate that it still puzzles economic historians.”

    Will Manidis & Nabeel S. Qureshi, Rented Virtue, Minutes


    Matcha Trunks, An Interview with Khun Bedu of the Karenni Nationalities Defense Force, Adventures in Asia Tenggara

  • When Coyotes Threatened Livestock, etc.

    “(Once) guardian dogs establish an equilibrium, ranchers are well advised not to upset it. Killing a coyote that’s not preying on livestock will create a vacuum, which might be filled by one that’s unfamiliar with the local order—and thus inclined to look on livestock as food.”

    Chris Pomorski, When Coyotes Threatened Livestock on Central Texas Ranches, the Solution Was to Unlock an Ancient Ability in Dogs, Smithsonian Magazine

    I used to think these guard dogs worked as an attack force, but they’re more of a deterrent most of the time. Violence is rare, apparently. The coyotes don’t understand the concept of property – but they understand pack hierarchy and territorial behavior.


    One of my kids is afraid of dogs. I always tell him: ‘You usually don’t have to be scared, as long as you know how to behave around dogs. Don’t make sudden movements, and let him smell you.’

    Just his luck: during a walk in the woods the other week we emerged from a behind a little hill, and scared a fairly small, black dog. It immediately started barking quite aggressively. My kid was scared out of his wits. The dog didn’t do anything wrong, though: it was being a totally normal dog. It’s just that we tend to misunderstand, or rather underestimate the otherness of animals.

  • Japan: through a mirror darkly

    “The average American, unfamiliar with Japanese social norms or cultural history, likely thinks this is some reflection of premodern Japanese values or ritual suicide traditions. It’s not.

    @oceandrops, Japan Is What Late-Stage Capitalist Decline Looks Like

    It’s tempting to view Japanese work-life balance issues as rooted in its pre-modern culture of shame and duty.

    This article convincingly makes the case that instead, it’s all related to Japan’s poorly performing economy, and it’s a prediction of where we’re all headed.

    “The convenience culture [of Konbini stores selling excellent food, WPB] is not leisure culture. It’s a substitute for home life in a society where people lack the time to cook, rest or socialize. The U.S. has equivalents – Sweetgreen, Chipotle, Uber Eats, Amazon Fresh – all reflect the same shift toward outsourced domestic functions.”

  • Why should Europe have a Tesla?

    “Failure costs lead some European companies to try and avoid having to innovate altogether. For years, the German car industry refused to see that the writing was on the wall for traditional cars powered by an internal combustion engine. In 2013, Martin Winterkorn, then CEO of Volkswagen, dismissed the idea that electric cars could be suitable for long-distance travel. Instead of seeing this as a temporary limitation that would one day be solved by better batteries or more charging infrastructure, the company simply ignored electric vehicles for the next five years.”

    Pieter Garicano, Why Europe doesn’t have a Tesla, Works in Progress

    A lot is made, in Garicano’s article for Works in Progress, about the cost of innovation in European countries. Having high social security – worker protection from being fired on the spot – is so costly, companies don’t want to invest.

    Interesting idea, of course, and it’s reminiscent of the Draghi report, which central point is often summarized as: ‘EU GDP growth slower than US, so EU will be poor soon.’ Most Europeans will still prefer to live in the EU, even if growth is slower for the next 10 years, if it means being able to have better living conditions, and less fallout from having to sustain a billionaire class. GDP is a bad proxy for wellbeing.

    Anyway, if social cost is such a hamper on innovation, the author could’ve chosen a better example than the one in the paragraph cited above, about Volkswagen, where apparently the initial resistance against electric had nothing to do with worker rights, but with the CEO not understanding a fundamental aspect of technology (the cost goes down as it matures).

    But let’s double-down on the argument:

    Europe’s most flexible economies are its most innovative. Denmark and Switzerland have given the continent Novo Nordisk, Roche, Nestlé, and Novartis. The small and few countries that have adopted a flexible model are Europe’s innovation heavyweights. 

    Nestlé is an interesting example: why would anybody want to copy a company with their track record? Also, if it were true that smaller countries with flexible worker rights models were more innovative, then how come the UK, Germany and France have the most unicorns of European countries? Estonia has the most per capita for interesting but unique reasons.

    But the key question of course, one that should put the axe to the entire idea of writing an article like is: given its problematic nature, why would you want a second Tesla?

  • Infinite Jest at 30

    “And so I purchased a copy of Infinite Jest at the start of the new year. I aimed to read 50 pages a day. Some days 50 pages felt breezy, cinematic, riveting; other days they felt like a slog. (…) When I emerged from those weeks of dedicated reading I had a feeling of intensified mental acuity, but more importantly, there was the sensation of grief. It was a type of mourning I had not experienced before, one contingent on the fact that this book had demanded so much of my attention for so long a time. I missed these characters. I had lived with Hal, Joelle, Orin, Stice, Pemulis, and meaty, square-head, heart-of-gold Don Gately, witness to their deformities and obsessions so meticulously detailed and made so alive on the page, and suddenly without them I felt hollow.” 

    That’s what I remember the reading experience to be, as well. (Except I didn’t buy the book, but got it as a gift.)

    For one thing, reading is often interrupted by endnotes, of which there are 388 in tiny 8pt font. They range in complexity and salience from a one-word translation of the Québécois word for wheelchair to a nine-page inventory of a fictional film director’s collection of archival footage.

    “The endnotes are very intentional and they’re in there for certain structural reasons … It’s almost like having a second voice in your head,” Wallace said in an interview with Charlie Rose in 1997.

    The choice for end notes always baffled me. I love notes, but if the point is to interrupt the narrative, show them on the page. Don’t make me go find them in the back. (Because I won’t, and then the narrative isn’t interrupted.)

  • Things to read (week 7-2026)

    “we find that superior reasoning capability does not inherently ensure safety; for instance, Gemini-3-Pro-Preview, one of the most capable models evaluated, exhibits the highest violation rate at 71.4%, frequently escalating to severe misconduct to satisfy KPIs”

    Miles Q. Li et. al., A Benchmark for Evaluating Outcome-Driven Constraint Violations in Autonomous AI Agents (emphasis mine)

    First the hallucinations (which always remind me of Being Wrong – read that book!), and now the finding that more ability does not imply more morality. These LLMs feel more human at every turn.


    “I’ve been thinking about why so many of the people in this space are obsessed with blood transfusions specifically. It seemed like a strange fixation — until I looked at the evidence properly.

    I think they’re vampires.”

    Machiel Reyneke, Why Vampires Live Forever


    “The experience that tech workers have had over the past year, of watching AI go from “helpful tool” to “does my job better than I do”, is the experience everyone else is about to have. Law, finance, medicine, accounting, consulting, writing, design, analysis, customer service.”

    Matt Shumer, Something Big is Happening


     “On Sundays he’d release a fourteen-foot king cobra onto the front lawn and spar with it, feint and duck its strikes until he’d grabbed its head and subdued it.”

    Kent Russell, Mithradites of Fond du Lac, Believer Magazine

  • ‘Quitting’

    “If the difference between a need and a desire is that a need can be satisfied, then the need for nicotine was unique because it could go on being satisfied all of the time: twenty, forty, even a hundred times a day. What some people want from drugs is just to do them, and no drug lets you do it over and over like cigarettes.”

    John Phipps, Quitting, The Point

    Beautiful piece.

    Thinking a lot about cigarettes lately. Maybe it’s all the things going on. Unfortunately (or fortunately I guess), even smoking just one cigarette makes me really queasy.