Roblog

Stay classy: mummy’s favourite (→ lrb.co.uk)

Tue, 10 Mar 2026 10:41:36 +0000

Andrew O’Hagan in the LRB reviews Andrew Lownie’s new book Entitled: The Rise and Fall of the House of York alongside Virginia Giuffre’s autobiography, and in doing so eviscerates Andrew Mountbatten-Windsor and the cheap, tawdry corruption with which he has coasted through his adult life:

“With some people, money and sex are the only truths. It’s the ultimate delinquency to believe that gratification itself is power. Bringing down the royal family may be the least terrible consequence of everything Andrew has done. When a lazy aristocrat from a dying dynasty uses a helicopter to travel seventeen miles, the edifice shakes. But when that same man rapes a 17-year-old and calls her a liar, it is the end of days. The gift of Andy and Fergie, which comes at too high a price, has been to bring the antiseptic of daylight to the culture of royal privilege.”

AI won’t automatically make legal services cheaper (→ normaltech.ai)

Sun, 15 Feb 2026 11:46:50 +0000

In a lengthy article that is my new gold standard for analysing the actual consequences of introducing AI into an industry, Justin Curl, Sayash Kapoor, and Arvind Narayanan explore the effects of AI on legal work.

Sayash and Arvind had previously authored AI as Normal Technology (also well worth reading), which argues that it’s a mistake by both boosters and doomsters to treat AI as some special category of thing that will lead to humanlike or superhuman intelligence; instead, we should treat it, and should remain in control of it, like any other technology. In particular, we should measure its impact by how it diffuses through society and industry, not by what it’s capable of in isolation, and we should expect that diffusion to take decades, not months.

This article is their attempt to apply the “AI as normal technology” thinking to a particular industry.

As a document- and language-focused industry, law has seemed ripe for LLM-based disruption; as an industry that often commands extremely high fees for its work, many people are desperate for that disruption to happen, and for legal work to become cheaper. That motivated reasoning is perhaps one of the reasons why law regularly tops the panic-inducing lists of “jobs that won’t exist in the future because of AI”.

The authors’ conclusions are much more nuanced. They look at regulatory concerns, which might inhibit AI’s progress into the industry; they look the fundamental dynamics of the industry, in this case the adversarial nature of common-law countries; they look at the impact of past technologies, which often failed to deliver the commodification of law that many people imagined.

AI, they conclude, won’t fix any of the structural problems faced by the legal industry and those who interact with it as clients, and may even magnify them; but it may be that the mere threat of AI disruption can be an external impetus to reform, reform that otherwise might have faltered through lack of will and coordination.

I think both the specific conclusions about law and the general framework outlined by the authors are enormously helpful contributions to the discourse around AI, and I’ll be trying to think along similar lines in my own work. #

How the New York Times uses a custom AI tool to track the “manosphere” (→ niemanlab.org)

Sun, 15 Feb 2026 09:44:35 +0000

The New York Times has developed a tool to download, transcribe, and summarise various right-wing podcasts, part of what they call the “manosphere”, in order to spot signs of division and discontent within Donald Trump’s base:

“When one of the shows publishes a new episode, the tool automatically downloads it, transcribes it, and summarizes the transcript. Every 24 hours the tool collates those summaries and generates a meta-summary with shared talking points and other notable daily trends. The final report is automatically emailed to journalists each morning at 8 a.m. ET. Currently, the tool is used by nearly 40 reporters across the newsroom.”

It’s a fascinating use of LLMs in the newsroom. Whether this ends up making bias worse, as the apparent consensus of grifters and racists becomes a signal that influences the Times’ own reporting, is a complicated but vital question. But this specific use-case is just one of many. The tool grew out of another, called Cheatsheet, which sounds essentially like a no-code tool that allows journalists to execute LLM-based workflows against custom data, and which has already led to significant investigative breakthroughs:

“The Initiatives Team started trialing other applications of LLMs to process large, messy datasets and file dumps on a case-by-case basis. Today, many of those live in a single spreadsheet-based tool. Reporters can drop datasets into Cheatsheet and then run different preset scripts and prompts. Each capability in the menu is known as a ‘recipe.’ Some of those recipes, like transcribing thousands of hours of video footage and summarizing transcriptions, are foundational to the Manosphere Report.

“Still in its beta, Cheatsheet has already been tested on about 300 users in the newsroom, with 50 of those being ‘really active users’, according to Seward. Right now, at least one new project is created in Cheatsheet every day. The tool has been used to investigate an election-interference group, to transcribe and translate Syrian prison records, and to find recent instances of Trump talking about Jan. 6. At times, Cheatsheet has even been used to take on more thorough historical analysis of podcasts.”

Saint Cavish at Gem Garden (→ youtu.be)

Sat, 14 Feb 2026 08:55:23 +0000

One of my favourite YouTube channels at the moment is @saintcavish, Christopher St. Cavish’s explorations of Chinese cooking. Christopher is no mere visitor, dipping in briefly and superficially to a vast and almost-unknowable cuisine; he has lived in Shanghai for over twenty years, and so is able to delve deep into culinary traditions and gain access to remarkable places, like this visit to a Michelin-starred fine-dining restaurant in Shenzhen.

The production values are perfect. There are no showy Netflix-style visuals or overly dramatic score. It’s just exceptionally clean, well-graded footage – shot by a single camera operator, Graeme Kennedy – of remarkable chefs cooking remarkable food, all with St. Cavish’s thoughtful commentary. Not one to watch while hungry. #

Elizabeth Goodspeed on imperfection as design strategy (→ itsnicethat.com)

Fri, 13 Feb 2026 12:44:50 +0000

Creatives’ social media accounts are awash with lo-fi, analogue aesthetics, most of which are created digitally, often with cookie-cutter kits that undermine the whole idea of what “analogue” is supposed to represent. Elizabeth Goodspeed understands why:

“The practical reality is that most people no longer have the time, tools, or support to make fully analogue work, even if they want to. The creative infrastructure that would make it viable – materials access, slower timelines, financial stability – isn’t widely available. Designers and illustrators are stuck in a bind: analogue signals value, but digital is what’s feasible. The result is a kind of strategic mimicry. The market is looking for particular cues, and designers have to find a way to hit them. It doesn’t help that glossy, computer-made work can now be mistaken for AI either; clean, high-fidelity digital craft has become suspect by default, making handmade a safer choice. You can think of adding in fake ink splatters a bit like penciling in a beauty mark: an intentional imperfection done to signal authenticity, rather than the byproduct of a real nuisance.”

Rather than using these analogue cues for merely surface-level styling reasons, or to signal “this isn’t AI”, Elizabeth hopes that we can use this analogue fixation as part of a broader reckoning about how and why creative work is made, and to whose benefit:

“When analogue collapses into surface style, it stops applying pressure to how work is made and valued. But it doesn’t have to be that way. The Arts & Crafts movement, for instance, emerged alongside mass mechanisation and responded not with nostalgia, but with structural reorganisation. Designers and makers pushed back against the factory’s division of labour by reasserting continuity between thinking and making. Objects were produced slowly, often collaboratively, with an emphasis on material knowledge and visible decision-making. They were sold through guilds and exhibitions that foregrounded craft as labour, not just aesthetic, and delivered tangible financial benefits to the people who made them. There’s a version of today’s analogue fixation that could move in this direction.”

The Mundanity of Excellence (→ academics.hamilton.edu)

Sun, 08 Feb 2026 21:26:18 +0000

Via Matt Webb, from his broader (and fascinating) post about the nature of work and what the roboticisation of domestic labour might take from us, comes this topical and counter-intuitive nugget about Olympians. How is it that they bring themselves to do the presumably mad levels of hard work and training that are required to get to the tops of their games? Well, because they enjoy it:

“The very features of the sport that the ‘C’ swimmer finds unpleasant, the top level swimmer enjoys. What others see as boring – swimming back and forth over a black line for two hours, say – they find peaceful, even meditative, often challenging, or therapeutic. … It is incorrect to believe that top athletes suffer great sacrifices to achieve their goals. Often, they don’t see what they do as sacrificial at all. They like it.”

It’s from a 1989 ethnographic study on Olympic swimmers by Daniel Chambliss. The central premise is that levels of sports are discontinuous, and that, while putting in more and more effort can make you better within a particular level of performance, it isn’t what progresses you to the highest heights:

“Having seen that ‘more is better’ within local situations, we tend to extrapolate: if I work this hard to get to my level, how hard must Olympic swimmers work? If I sacrifice this much to qualify for the State Championships, how much must they sacrifice? We believe, extrapolating from what we learn about success at our own level, that they must work unbelievably hard, must feel incredible pressure, must sacrifice more and more to become successful. Assuming implicitly that stratification in sports is continuous rather than discrete (that the differences are quantitative) we believe that top athletes do unbelievable things. In short, we believe that they must be superhuman.”

Neither talent nor hard work are good explanations for their success. Instead, Chambliss argues, the cause is more qualitative than quantitative; the extremely successful do different things rather than more things, and keep doing them to a habitual and mundane degree.

“After three years of field work with world-class swimmers… I wrote a draft of some book chapters, full of stories about swimmers, and I showed it to a friend. ‘You need to jazz it up,’ he said. ‘You need to make these people more interesting. The analysis is nice, but except for the fact that these are good swimmers, there isn’t much else exciting to say about them as individuals.’ He was right, of course. What these athletes do was rather interesting, but the people themselves were only fast swimmers, who did the particular things one does to swim fast. It is all very mundane. When my friend said that they weren’t exciting, my best answer could only be, simply put: that’s the point.”

Was the green lantern a good chap? (→ open.substack.com)

Tue, 20 Jan 2026 10:16:49 +0000

Dan Davies has written about how many systems (including the British constitution) use what he terms a “good chap” regulatory system: i.e. they rely on the system being mostly populated by decent people who obey the rules, even though they don’t technically have to.

That has obvious flaws, like when someone (Boris Johnson, say, or more disastrously Donald Trump) comes along who decides not to be a good chap. But, Dan argues, there’s a certain advantage to the naïveté:

“It is always possible to break norms, if you’re really determined to. Everything does, in fact, depend on having people in positions of power who respect the rules of the game. The British ‘good chap’ system is just much more blatantly in your face about it.

“Which might account for the longevity of the Westminster system. It is incredibly fragile, but it’s obviously fragile, and in this way achieves a sort of paradoxical antifragility. In a ‘good chap’ system, when a bad chap shows up, all the good chaps know that they have to band together and oppose or get rid of them, because they know that there are no systemic constraints on badchappery. In a system that’s meant to be full of checks and balances, it is much easier for a kind of bystander effect to develop, where everyone waits for the system to protect itself without understanding that the system is just them.”

The contrast with the US in its current mode is striking; I definitely see that bystander effect, where people continually expect some kind of check or balance to take effect. It certainly hasn’t yet. #

It’s hard to justify Tahoe icons (→ tonsky.me)

Tue, 06 Jan 2026 10:22:26 +0000

A brilliant takedown of the menu icons in the latest version of MacOS, a version I’m still steadfastly refusing to upgrade to, from Niki Prokopov.

Apple decided to do a thing that was both impossible and undesirable, then did it badly:

“In my opinion, Apple took on an impossible task: to add an icon to every menu item. There are just not enough good metaphors to do something like that.

“But even if there were, the premise itself is questionable: if everything has an icon, it doesn’t mean users will find what they are looking for faster.

“And even if the premise was solid, I still wish I could say: they did the best they could, given the goal. But that’s not true either: they did a poor job consistently applying the metaphors and designing the icons themselves.”

In referencing the Human Interface Guidelines from 1992, Niki makes a point that many people would do well to remember:

“…is an interface manual from 1992 still relevant today? Haven’t computers changed so much that entirely new principles, designs, and idioms apply?

“Yes and no. Of course, advice on how to adapt your icons to black-and-white displays is obsolete. But the principles – as long as they are good principles – still apply, because they are based on how humans work, not how computers work.

“Humans don’t get a new release every year. Our memory doesn’t double. Our eyesight doesn’t become sharper. Attention works the same way it always has. Visual recognition, motor skills – all of this is exactly as it was in 1992.”

Jevons only knows

Wed, 10 Dec 2025 15:00:00 +0000

In mid-19th century Britain, deep in the throes of the industrial revolution, there was a pressing question: there was only so much coal in the world, so how did you avoid running out of it? Coal was what had made industry possible, “the material energy of the country, the universal aid, the factor in everything we do.” Running out would be a total disaster.

Many people thought the answer was to make the things that used coal more efficient, so they could achieve the same output while using less coal. By doing that, they thought, you’d ease the demand for coal, and allow the limited reserves to last for longer.

In his 1865 book The Coal Question, the economist William Stanley Jevons took the opposite view. Increasing the efficiency of coal wouldn’t reduce demand for it, Jevons said; in fact, it would have the opposite effect. If you ran a factory on coal, and some new technology made your factory significantly more efficient, you wouldn’t respond by making the same amount of stuff in less time; you’d make more stuff, more cheaply, and use more coal in the process. If you made coal more efficient in the hope that people would use less of it, Jevons said, you were in for a nasty surprise.

This is the Jevons paradox: a situation where an increase in efficiency of something creates so much more demand for it that any effect of the increased efficiency is wiped out. It requires:

The efficiency of the thing to be capable of being improved as the result of new technology
Improvements in efficiency to lead to lower prices for the consumer (i.e. no monopolies or general supplier bargaining power that means suppliers capture the surplus)
Lower prices to lead to much greater demand from the consumer (i.e. demand to be highly elastic)

Coal is the textbook example, but think of computing. Computers are thousands of times more powerful – per watt of power or dollar of cost – now than they were in, say, 1990. But we haven’t responded to that improvement by simply running the same computer programs as we did then, only hundreds of times more quickly and cheaply. Instead, many more possibilities for computerisation have been opened up by the fact that computing power is now much cheaper. We’ve put computers into fridges, wristwatches, cars and children’s toys; we’ve computerised millions of processes that previously took place in person or on paper; we’ve found countless little corners of the economy that we can shove computers into. We use far more computing power nowadays than in 1990 precisely because computers are more efficient now. Conditions 1, 2 and 3 apply; it’s the Jevons paradox.

On the other hand, you might think of the farming industry. In the UK in 1860, agriculture employed nearly a quarter of the population. Since then, it’s become much, much more productive: feeding the country in 2025 requires us to employ only about one per cent of the population in agriculture, and food in general is much cheaper in real terms. But what we haven’t seen is an explosion of demand for this cheaper, more efficiently produced food. It turns out that demand for food is pretty inelastic; in developed countries we do tend to eat too much, but we don’t respond to a significant decrease in the price of food by eating much more of it. Conditions 1 and 2 hold, but not condition 3. And so while agriculture was 20% of UK GDP in 1860, it was 7% by 1913 and is now just 0.6%.

The Jevons paradox has been enjoying some time in the spotlight in the last couple of years, thanks to our old friend AI. It’s sometimes used with reference to AI itself (i.e. that as AI becomes cheaper, we’ll use more of it). But lots of people, many of them the executives of AI companies, suggest that the Jevons paradox will apply to industries affected by AI: that AI will increase efficiency, but that this will lead to industries growing rather than being destroyed.

You can see how this could be either wishful thinking or a cynical distraction technique on the part of AI executives. But for those of us outside the AI industry itself, this line of reasoning suggests a pressing question. Could it potentially be true of your industry, and what would that mean for you? Is your particular industry likely to end up like computing, or like agriculture? And if it does look like computing, and so is subject to the Jevons paradox, what does that mean for employment within the industry – rather than just overall demand and output? After all, things might grow significantly and create a rosy picture for GDP, but without the need for pesky human beings.

The criteria that matter are:

Could AI lead to the Jevons paradox in this industry? That is, could AI potentially increase efficiency in a way that lowers prices and unleashes previously suppressed demand, so that overall demand increases?
Will there be a significant role for humans left in the post-AI version of this industry? Will humans be merely augmented by AI, or replaced by it?

The answers to those questions implies a 2×2. One axis reflects what happens to demand for a given industry, and the other reflects what happens to the human beings working in it:

In the bottom-left quadrant, the industry remains on its existing growth trajectory – there’s no unleashing of huge demand – but the number of humans employed in it dwindles towards zero. This is what happened to agriculture in the past. I can imagine this happening to all manner of routine knowledge work jobs in the future: compliance checking, review of legal documents, payroll processing, quality control in manufacturing. There is no suppressed demand for these things, but they are automatable; they will therefore be automated down to pure commodities by AI.

In the bottom right, demand also remains on its pre-AI trajectory, but the story is slightly rosier for humans; AI, perhaps through automating a previous bottleneck, makes them more efficient, but they remain needed. They can do more, but their ability to do more doesn’t create increased demand for what they do. An example of this might be radiology: the radiologist Ben White has written about his discipline in this context. AI might well make radiologists faster at interpreting scans. But it’s unclear that this would translate into lower prices, given how the healthcare industry works, and even if it did, demand for radiology is driven by illness rather than price. There is no Jevons effect here.

In the top left, we see the future that AI execs perhaps dream of: widespread automation that replaces humans en masse, vastly increasing the demand for that industry’s output. Think the translation of commodity content into different languages, where costs were too high to pay a human to translate it but where things can be translated automatically and on the fly with AI. Or lots of call centre jobs, where there is a demand for 24/7 multilingual customer support that can’t currently be met, but potentially could in a world of AI. Or tons of things that were previously the preserve of a marketing or creative agency: creating bits of copy or little graphics.

And in the top right, we find AI that augments what humans do, increasing the demand for it while keeping them in jobs. To me, software development seems the obvious industry that fits this shape: we’ve been supply-constrained on software for decades, and there are zillions of places where software could improve efficiency or add value that it thus far hasn’t been able to reach, because there aren’t enough software developers and what they do is too expensive. Demand for software is effectively infinite. Another example is pharmaceuticals research, where a human scientist needs to design and oversee trials but where AI could potentially screen compounds or suggest candidates at a rate far faster than a human could. Or business analytics, where currently the human analysis time is a significant bottleneck, where there’s no realistic limit to the amount of analysis a business would create and consume if its analysts had the capacity to produce it, but where human judgement is still essential to making good decisions.

This, then, feels like a useful framework for decision-making. It doesn’t require you to believe that AI can transform industries in its current state. But it lets you think about… what if?

As an employee, you might begin to worry if you’re in a role or industry on the left side of the chart, where humans will be replaced wholesale. The bottom left will become a particularly grim place to be. But it also leaves complicated questions for those working in industries on the bottom right. What view do you take about the growth potential of those in the top right compared to those in the bottom right? Will the top-right industries race away, growing much faster? Do you risk being stuck in a dead-end industry, with the best outcome being the capturing of obsolescence rents as part of a smaller and older workforce, the future equivalent of today’s COBOL programmers?

For business owners, the bottom left quadrant is scary; a race to the bottom against low-cost providers. The bottom right is concerning for the same reason as it is for employees: the risk of long-term stagnation. But the top left is a capitalists’ dream; lots of growth and no pesky humans to slow things down. Expect lots of investor attention on that quadrant, for good and for ill.

That leaves the top right. It’s an exciting place, that lots of people want to play in, but even here there are pitfalls. Can you adapt to change as quickly as you need to? Is the AI-augmented version of your role something that’s still fulfilling, or are you reduced to a decerebrate AI overseer, forced to work at the pace of a machine, a “reverse centaur” as Cory Doctorow puts it? Does AI rip out the craft from what you do, rather than feeling like a superpower? As either a business owner or an employee, are you able to ride the wave of growth rather than being swept away by it?

Giving names to the quadrants sums up roughly how I’m feeling about them, looking ahead to a potentially AI-disrupted future:

Jevons, for what it’s worth, was right about coal. British coal consumption quintupled in the fifty years after he wrote The Coal Question. But the miners themselves, and the environment, paid a terrible price for it. I’m not sure whether I’m right about the industries I work in, but I’m sure as hell hoping I’m able to stay in the top right – and that the future is rosier for them than it was for coal. #

Chain Your Heart to a Star: Fee Warner and the Marc Bolan Rock Shrine (→ youtu.be)

Mon, 08 Dec 2025 13:05:24 +0000

A couple of months ago I made my first documentary film: Chain Your Heart to a Star. It’s the story of Fee Warner, who for 48 years has been visiting the site of Marc Bolan’s death in south-west London. There she has built, and continues to maintain, what has become a shrine to Bolan and T. Rex. The shrine is visited by people from all over the world, and is enduring evidence of the depth of feeling we humans have for music and musicians. #