In February 2022, while he was plugging away at rendering the US writer Dana Spiotta’s novel Wayward into French, the literary translator Yoann Gentric decided he needed a bit of light relief. He would test whether AI could put him out of work.

Gentric had been grappling with a short non-verbal sentence that described the book’s protagonist’s feelings upon opening a window: “Bright, sharp night air, bracing.” He put the prompt into DeepL, a neural-network-powered machine translation engine that regularly outperforms Google Translate in accuracy assessments.

The proposed translation was reassuring, with his job security in mind: L’air de la nuit, vif et vif, était vivifiant (The night air, lively and lively, was enlivening.) AI had translated the sentence’s meaning but was seemingly unaware that the repetitions rendered the line absurd. It was far inferior to his own translation that would be published in the book a year later: L’air pur et piquant de la nuit, vivifiant.

When Gentric repeated his experiment this spring, however, the outcome made him feel less at ease: L’air nocturne était vif, pur et vivifiant, DeepL suggested this time. The online translator still lost the sentence’s stylistic trait by adding a verb, but it had learned to use three different words that even had a musical ring to them. “I don’t know if it’s just chance or a fine-tuned algorithm at work, but nocturne and pur is not bad,” said Gentric.

Chatbots running on large language models (LLMs) – neural networks trained on vast amounts of text to generate natural-sounding language – are rapidly infiltrating every aspect of our work and leisure lives. But few professional sectors are being disrupted by the technology as rapidly as the translation industry in Europe, home to more than 200 languages and a booming tech sector.

According to a recent joint survey by the French authors’ societies ADAGP and the Société des Gens de Lettres, 79% of translators believe the rise of AI “poses a threat of replacing all or part of their work”. In Britain, a 2025 survey found that 84% of translators questioned expected lower demand for human translation, resulting in lower pay.

Those fears concern the future, but for many translators the nature of their work has already changed. Laura Radosh, a Berlin-based German-to-English translator, used to get about four job requests per month from clients including universities, professors and museums. Last year, the number of offers dropped to one each month.

Many of them were “post-editing” jobs, which required her to correct texts that had already been run through a machine-translation engine. “Post-editing took me as much time as translating from scratch,” said Radosh.

Far less creatively fulfilling than translating from scratch, post-editing is also less well-paid: usually compensated by the hour rather than by the page or by the book, it is paid “at unacceptable rates considering the work involved”, according to the French translators’ association. In Germany, publishers have been found to offer typical rates of two to eight euros per page – a quarter of the average pay for translating a page from scratch.

But rates for regular technical translations have tumbled too. “I got offered a job at 60 cent[s] a line,” said Radosh. “Before then, 80 cent[s] was the lowest rate I had ever come across.”

Even before the advent of LLMs, translation was a precarious profession: a recent survey by the German translators association VdÜ found that the average income for literary translators – traditionally at the lower-paid end of the sector – was as little as €20,363 euros per annum before tax. But the latest changes in the industry mean that for many translators, the numbers no longer stack up – Radosh recently took a part-time job doing book-keeping for an NGO.

Marco Trombetti, the co-founder and CEO of the machine translation company Translated, said: “Without help, the human brain basically is able to produce about 3,000 words a day of translation. Beginners will manage about 1,500, the best translator in the world may manage 6,000, but the variation is not that big.”

The cost of human translation, he argued, had until now been defined by the number of neurons we have in the brain. “That’s around 100bn,” Trombetti said. “But if we change that, then we change the unit economics of translation.”

Yet the speed of technological change is also revealing what human translators still do best. For one, many machine translators still struggle with context. The German-British academic publisher Springer Nature offers its authors the option to have their books auto-translated into other languages for free, but in spite of assurances of subsequent “human checks”, this process has in the past led to comical results.

In 2024, Springer Nature machine-translated into German an English-language book by a group of Indian academics called ‘Capital’ in the East: Reflections on Marx. In the chapter headings, however, the machine translator DeepL had rendered “capital” not as Kapital in the intended sense, but Hauptstadt, meaning “capital city”.

A spokesperson for Springer Nature said in a statement: “Our AI‑supported translation is human‑led and reviewed by professional editors. Errors like this are rare and regrettable, and this instance relates to a limited pilot that has since ended.”

Jörn Cambreleng, the director of Atlas, a French organisation promoting literary translation, said: “Machine translation is not creative. These systems are built to produce sentences that are generic, sentences that have been said before or sound like they have been said before. Whereas good human translators strive to put into words something that has never been said before.”

One of the ironies of the upheaval is that literary translation now appears to be a comparatively safer career choice than its technical counterpart.

The HarperCollins-owned imprint Harlequin France has confirmed that it is working with a French communications agency, Fluent Planet, to produce translations that are generated by AI software and then post-edited by humans, although for now such trial runs are confined to the pulpier reaches of the market: Harlequin’s titles include A Mistress’s Confession and The Embrace of a Prince.

In Germany, where the total number of new published books has been gradually declining year on year, literature in translation has held up remarkably well, with 8,765 books in translation published in 2024 making up a historically high 15% of the overall output. Increasingly, authors are also contractually obliging their publishers not to use AI in the translation process, said Marieke Heimburger, a Danish-to-German translator who chairs VdÜ.

“AI really cannot do dialogue,” said Katy Derbyshire, a Berlin-based translator who has rendered into English novels by Clemens Meyer, Christa Wolf and others. “When you are translating from scratch, you learn to understand the characters and their motivations, and you’re constantly adjusting them in your head – to individual situations, but also to genre. The dialogue that AI came up with just didn’t suit the character description at all.”

Being human helped the translation process, she added. “My body has experienced all the pain and the joy that literature strives to convey. I understand what someone might scream when they hit their toe on the bed frame – an algorithm doesn’t.”

Fernando Prieto Ramos, of the University of Geneva’s faculty of translation and interpreting, said his centre had noticed a drop in applications to translation courses three years ago, when the rise of generative AI fuelled the hype around machine translation. “But the trend is gradually reverting again with a more diversified training offer,” he said.

Even people who develop machine translation software concede there are tasks that remain beyond their product’s reach. “If in Italian I say Solo tre parole: non sei solo, then a literal translation into English would be ‘Just three words: you are not alone,’” said Trombetti, who founded Translated in 1999. “But you’ve ended up with four words, not three. That’s something that machine translation still struggles with.”

Heimburger said: “I am not really scared of AI, because I know it cannot do what I can do. What I am afraid of is the people who think that AI can do my job.”

In the last 15 years, a linked series of unprecedented technologies have changed the experience of personhood across most of the world. It is estimated that nearly 70% of the human population of the Earth currently possesses a smartphone, and these devices constitute about 95% of internet access-points on the planet. Globally, on average, people seem to spend close to half their waking hours looking at screens, and among young people in the rich world the number is a good deal higher than that.

History teaches that new technologies always make possible new forms of exploitation, and this basic fact has been spectacularly exemplified by the rise of society-scale digital platforms. It has been driven by a remarkable new way of extracting money from human beings: call it “human fracking”. Just as petroleum frackers pump high-pressure, high-volume detergents into the ground to force a little monetisable black gold to the surface, human frackers pump high-pressure, high-volume detergent into our faces (in the form of endless streams of addictive slop and maximally disruptive user-generated content), to force a slurry of human attention to the surface, where they can collect it, and take it to market.

Fracking (of the Earth and of our minds) produces tectonic instability, toxicity and the despoliation of our landscapes, natural and social. We now know that the heedless exploitation of our external environment has been so relentless and irresponsible that human survival on Earth has been placed in actual jeopardy. The new “gold rush” into the inner environment of the human psyche is well on its way to effecting parallel, if even more insidious destruction.

The stakes are existential. And that is because, rightly understood, our actual human “attention” – the thing the frackers want, in the form of our eyes on their screens – is nothing less than our ability to care, our ability to think, our ability to give our minds, time and senses to ourselves, the world and each other. To commodify that is to commodify our very beings. The problem isn’t “phones”, and it isn’t “social media”. The problem is human fracking, a world-spanning land-grab into human consciousness – which big tech is treating as a vast, unclaimed territory, ripe for sacking and empire.

That’s the bad news. The good news is that novel forms of exploitation produce novel forms of resistance. What fills the coffers of the six largest corporations on the planet is nothing other than the stuff of our humanity. Which is to say this new fight for our attention stands in a long line of clashes between those who are willing to reduce people (their labour, their eyeballs) to cash value and those who insist on a higher view of human flourishing. This history is long and complex, and often painful. But it tells us this much: we can fight back. Indeed, we must.

So what is to be done about this new kind of human exploitation that is harming us – harming children and adults, compromising our deliberative politics and our psychological wellbeing? Regulatory efforts are piecemeal and actively thwarted by the powerful interests in play. Psychopharmacological fixes for the ever-expanding damage merely monetise the destruction in a complementary way and render us better able to submit to conditions that are palpably at odds with human flourishing. How to confront a problem that is both unspeakably intimate and unthinkably extensive?

The answer is clear: we, the actual people of this planet, must come together in decisive solidarity; we must say no to the human frackers, and do so by insisting, in new ways, that human attention is human, and it is ours, and we will use it to make the worlds in which we want to live. In other words, we need a movement.

Think that sounds quixotic? Well, keep in mind that’s how actual change actually happens. And it can happen fast. The environmental movement as we know it did not exist in 1950, but by 1970 it was a global force. In 1946 Reynolds Tobacco was using doctors to promote cigarettes. Fewer than 20 years later, the American Medical Association and the US surgeon general publicly averred that smoking caused lung cancer.

And the changes get much bigger than that. Precious few do-gooders devoted themselves to environmental politics in 1925. That’s because “environmental politics” wasn’t even a thing. It took a cultural shift (and the work of advocates such as Rachel Carson) across the mid-20th century to establish the physical environment – the unity of land, water, and air that produces shared life – as a politically tractable object around which diverse groups could organise. This is to say that the very structures of politics, not just our beliefs and hopes, are themselves emergent forms. New things come into being, and old things pass away.

Where attention is concerned, there are mounting signs that we are reaching an inflection point. People of all sorts, Maga Republicans and Mamdani progressives, hipsters in Portland and evangelicals in Arkansas – people who don’t agree about anything – all actually do agree that something is totally wrong with a world in which everyone spends nearly all their time scrolling endlessly through the algorithmic feeds of their social media, a world where military-grade technology and trillion-dollar corporations take aim at children, and feed them whatever it takes to keep them hooked.

You can only abuse people so much, and then, eventually, they turn, they rise, they insist on something else. Already politicians on the right and the left are identifying this issue as one that moves the electorate. In 30 years, we will look back, and this era – the wild west of the tech princes’ smash-and-grab into our hearts, souls and relationships – will be difficult to explain to our grandkids. “How did you all let that happen?” they will ask. And we’ll have to say: “It’s hard to explain: it happened before we noticed; it was so much fun, especially at first; it took us time to figure out what was going on … ”

But we are figuring it out. We write as representatives of a fast-growing and increasingly well-organised movement, focused on pushing back against the human frackers, and giving shape to a new politics of human attention. At the heart of our efforts? The formation of broad coalitions devoted to the politics of human attention, the practice of diverse forms of study that call forth the life-giving powers of the mind and senses, and the promotion of sanctuary spaces for the protection and cultivation of the kinds of attention that make life good. We call this work attention activism.

Our claim? That all of us already have the tools to resist the frackers, because all of us already have things we do and care about that put us beyond the reach of the algorithms. We all already know the deepest truth: that true human attention isn’t the click and swipe of screen time. True human attention is love, curiosity, daydreaming and taking care of ourselves and others.

Yes, new technologies give rise to new kinds of exploitation and resistance. But new forms of exploitation can even give rise to genuinely new forms of politics. You couldn’t brutalise an industrial proletariat before the factory system. Steam engines set the conditions of possibility for this development. They weren’t themselves a “problem”, of course; they gleamed and were precise and powerful. Who could see them operate without awe? But they also created a world in which it was possible to aggregate and extract physical labour from human beings in a revolutionary way. Along the way, they created a new kind of political subject, Homo economicus, a person who had been reduced, in the calculus of modernity, to “labuor value”. Actual revolutions followed – and a new kind of politics was born which reflected a new world of industrial labour, and new forms of labour solidarity, such as unions and workers’ parties.

The new system of human fracking is turning all of us into attentional subjects in a powerful new way. Homo attentus is the end user of every networked system – economic, political, expressive. With this new form of life comes, as we have discovered, appalling new vulnerability. But we are on the brink of understanding the new power that has come into our hands in the fracklands. We believe a new kind of politics beckons. What will it look like? It is hard to say. And there are reasons to be fearful. But if we, the people, can take up the banner of a new kind of freedom movement – a movement for the true freedom of attention itself, what we call attensity – and deploy our truly human attention in new ways, with a new understanding of the stakes, we can defy the frackers, and insist on creating, together, a human world.

D Graham Burnett is professor of history at Princeton University. Alyssa Loh is a film-maker. Peter Schmidt is a writer and organiser. The authors are members of the Friends of Attention coalition, and co-editors of ATTENSITY! A Manifesto of the Attention Liberation Movement (Particular).

Further reading

The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness by Jonathan Haidt (Penguin, £10.99)

How to Do Nothing: Resisting the Attention Economy by Jenny Odell (Melville House, £14.99)

The Sirens’ Call: How Attention Became the World’s Most Endangered Resource by Chris Hayes (Scribe UK, £16.99)

On the street, in a huge line, undeterred by rain, they gathered to pass books.

Bibliophiles, builders from nearby construction sites, kids with their parents, all stood for hundreds of metres along Bourke Street in Melbourne’s CBD on Thursday morning in a human chain.

They were there to help the beloved bookshop, Hill of Content, move from its location of more than 100 years to a new home.

In wintry weather 300 people stood in a line passing thousands of books up Bourke Street, from Hill of Content’s old store into the new.

As rain started falling people pulled out umbrellas and raincoats, with the books wrapped in brown paper.

Hill of Content is the city’s oldest bookstore, opening in 1922 at 86 Bourke Street. The three-storey heritage-listed building it occupied for 103 years was sold for $5.3m last year, with the bookshop’s owners forced to start searching for a new home.

Diana Johnson, who owns Hill of Content with her husband, Duncan Johnson, said the human chain would pass 17,000 books up to the new store.

“There are a fair number of people I know in the line, lots of loyal customers, we are so grateful they have supported us all those years,” she said.

“We couldn’t possibly close the shop down on our watch. It’s been in Melbourne literacy for over 103 years. So we decided we would continue it on.”

Within an hour, the army of booklovers had already put hundreds of books on the new shelves, Johnson said.

She was inspired to try relocating the store’s inventory with a human chain after a customer told her of a “book brigade” in the United States that helped move Serendipity Books in Michigan to a new location.

“I am just so grateful,” Johnson said.

The new premises at 32 Bourke Street are just metres away from the old, and Hill of Content put out a call on social media earlier this month asking booklovers to help it move.

skip past newsletter promotion

after newsletter promotion

“Many hands make light work. Join us and see the power of bookish community,” the Instagram post read.

On Thursday, literature lovers turned out in force.

Jess, who was standing in line, described it as “a human conveyor belt”.

She admitted she “didn’t come often” to the old store, but would visit the new one.

A little further up the chain, three young builders had jumped into the line after seeing it happening while on break from work.

“We were just doing the fit-out work on the building, next minute people lining up, so we thought we’d get down,” Wyatt said.

Standing next to them, Angela joked they were getting “book fit” as they stood in line.

“These young men, they don’t really read a lot, I don’t think,” she said. “But we’ve been giving them some recs.”

In line, there was also a lot of discussion about whether it is Hill of Content, as in satisfied, or content, as in production – a debate that has cropped up among the city’s booklovers.

“It’s what you want it to be,” said Johnson, before adding “but it’s actually hill of content”, like the peaceful.

“When Albert Spencer first came down here and set it up in 1922 it was a very, very wild area,” she said.

“So he went up into the gardens and walked around, wondering, what can I call my shop.

“[He called it] the hill of content, where you can come in and you can be contented and get away from all that’s happening outside. It’s been 103 years.”

Around 300 people gathered outside Melbourne’s oldest bookshop, Hill of Content, to help it move to its new location just 130 metres down the road. The store first opening in 1922 and the three-storey, heritage-listed building it occupied for 103 years was sold for $5.3m last year, with the bookshop’s owners forced to start searching for a new home. Diana Johnson, who owns Hill of Content with her husband, Duncan Johnson, said the human chain would pass 17,000 books up to the new store

Hundreds form human chain to help Melbourne’s oldest bookshop relocate after more than a century

Residents of all ages in a small Michigan community formed a human chain and helped a local bookshop move each of its 9,100 books – one by one – to a new storefront about a block away.

The “book brigade” of about 300 people stood in two lines running along a sidewalk in downtown Chelsea on Sunday, passing each title from Serendipity Books’ former location directly to the correct shelves in the new building, down the block and around the corner on Main Street.

“It was a practical way to move the books, but it also was a way for everybody to have a part,” Michelle Tuplin, the store’s owner, said. “As people passed the books along, they said ‘I have not read this’ and ‘that’s a good one’.”

Momentum had been building since Tuplin announced the move in January.

“It became so buzzy in town. So many people wanted to help,” she said on Tuesday.

Tuplin said the endeavour took just under two hours – much shorter than hiring a moving company to box and unbox the thousands of titles. The brigade even put the books back on the shelves in alphabetical order.

Now Tuplin hopes to have the new location open within two weeks.

The bookstore has been in Chelsea, about 60 miles (95km) west of Detroit, since 1997. Tuplin has been the owner since 2017 and has three part-time employees.

About 5,300 people call Chelsea home and residents described it as a place where neighbours help neighbours.

“It’s a small town and people just really look out for each other,” said Kaci Friss, 32, who grew up in Chelsea and has worked at the bookstore for a little over a year. “Anywhere you go, you are going to run into someone you know or who knows you, and is going to ask you about your day.”

Friss said Sunday’s book brigade reminded her of “how special this community is”.

Amanda Palmer, the musician and former partner of Neil Gaiman, has denied allegations of human trafficking and negligence made in a civil lawsuit filed by a woman who previously worked for the couple in New Zealand.

“I will not respond to the specific allegations being made against me except to say that I deny the allegations and will respond in due course,” wrote Palmer in an Instagram post on Friday, adding that her “heart goes out to all survivors”.

On 3 February, Scarlett Pavlovich filed a lawsuit to district courts in Wisconsin, New York and Massachusetts accusing Gaiman of repeated rape and sexual assault, and Palmer of “procuring and presenting” her to Gaiman “for such abuse”.

Nine women have now accused Gaiman of sexual misconduct, eight of whom were interviewed for a New York Magazine piece published on 13 January.

The following day, Gaiman published a statement on his website stating that he had “never engaged in non-consensual sexual activity with anyone. Ever.”

Palmer and Gaiman, who married in 2011 and had a child together in 2015, announced in 2022 that they had decided to divorce.

On 15 January, Palmer said in an initial Instagram statement that because of ongoing custody and divorce proceedings, she was unable to offer public comment on the allegations. A representative for Palmer told NME that she “is profoundly disturbed” by the allegations against Gaiman.

The lawsuit states that Pavlovich met Palmer in Auckland in 2020, when Pavlovich was 22, and the two became acquaintances. Pavlovich would sometimes run errands for Palmer, and eventually became the couple’s nanny.

The lawsuit alleges that Gaiman repeatedly sexually assaulted Pavlovich while she was working without pay during a “three-week indenture”. At the time, she was “broke and homeless”, and Gaiman and Palmer “intentionally withheld” pay to keep her “trapped, vulnerable, and penniless”, it claims.

Pavlovich said when she went to Palmer about the assaults, Palmer told her that other women had previously come to her about abusive sexual encounters with Gaiman.

The lawsuit states that Palmer “knowingly approached and procured the services of Scarlett with reckless disregard for the fact that Gaiman would force Scarlett to engage in commercial sex acts” with him.

“I thank you all deeply for continuing to respect my recent request for privacy as I navigate this extremely difficult moment”, Palmer wrote in her post on Friday. “I must protect my young child and his right to privacy.”

Eric Hazan, a lifelong Parisian who died in June, wrote several books about his hometown, with a particular focus on the class politics of the built environment. In Balzac’s Paris he revisits the 19th-century social geography of the French capital through the fiction of one of its most famous novelists. Honoré de Balzac’s La Comédie humaine (Human Comedy) – a vast series of novels and stories depicting French society between 1814 and 1848 – is one of the canonical texts of literary realism. In these works, Hazan writes, the street is more than just a setting: “The places where the characters live and evolve are part of their personality; they define them in the same way as their physique, their dress or their psychology.”

First published in France in 2018 and now available in English thanks to David Fernbach’s translation, Balzac’s Paris is a blend of literary criticism and historical psychogeography. Hazan narrates in the manner of a tour guide, hopping from location to location and offering up nuggets of commentary: pertinent quotes from the novels, or Balzac’s personal correspondence; etymological titbits; an apposite line from Baudelaire or Proust. The format, and the languid, dizzyingly directionless prose style, will be familiar to readers of Hazan’s best-known work, The Invention of Paris, a sprawling radical history of the city, which was published in English in 2010.

One moment we’re amid the jumbled streets and medieval architecture of Old Paris, where Balzac’s characters can be found frequenting the gambling dens of the Palais-Royal, or schmoozing at the Opéra. The next we’re being led through New Paris, the area stretching from Montmartre to the city walls, which was built up during a flurry of construction under the July Monarchy (1830-48). Its residents range from the wealthy bankers of the fashionable Chaussée-d’Antin district to the sex workers, known as lorettes, associated with the neighbourhood around the church of Notre-Dame-de-Lorette. (Hazan observes that Balzac portrays them sympathetically, “whereas the noble ladies of high society are either seductive, egotistical, and brutal … or else more or less virtuous dimwits”.)

The less salubrious districts are home to a number of socially marginalised characters. In the Latin Quarter we meet an escaped convict, a mentally fragile marquis down on his luck, and the slimy title character of Gobseck (1830), a moneylender who brags: “I like to leave mud on a rich man’s carpet; it is not petty spite, I like to make them feel a touch of the claws of Necessity.”

For the most part, however, Balzac – a staunch conservative and monarchist – was concerned with depicting the comings and goings of high society. Across the novels, we meet workers “only in passing … because Balzac’s characters, whether bourgeois or aristocratic, have no business in the working-class suburbs”. The political strife of the 1830s rarely featured: “nothing happened – at least nothing in the streets: neither riots nor uprisings, nor insurrections. It is just in the turn of a sentence, in a quick allusion, that we sometimes perceive the distant echo of battles.”

Hazan’s peregrinations culminate in a thoughtful disquisition on literary realism, in which he suggests that the term itself is misleading. He points out that Balzac’s novels make no mention of the railways that were then proliferating in the city, or the new fortifications built in the 1840s; people with chestnut hair – the majority – are conspicuously under-represented in the Human Comedy, as Balzac preferred to populate his stories with characters who have either blond or jet-black hair.

Were this any other writer, the appropriate response would be: so what? He was a novelist, not an archivist. But something about the panoramic breadth and descriptive detail of the Human Comedy, with its cast of about 2,500 characters, has tempted generations of readers and critics to view it as something akin to a factual chronicle. Of course, the novels were only intended as entertainment. “Balzac,” Hazan quips, “is no more of a realist than Scheherazade”. Even so, the mythos of great literature bleeds into our sense of history.

skip past newsletter promotion

after newsletter promotion

There is, it seems, an unwritten law in the world of artificial intelligence, which I will attempt to distil here: “Any discussion of AI must include an early and robust reference to the Terminator”. Though the 1984 James Cameron film and its 1991 sequel are quite good, here are two equally made-up but probably mostly true facts: no one under the age of 30 has seen either film and, in any case, neither film has anything particularly insightful to say about AI. But here we are, and the relentless analyses of the moment we are in – where we apparently stand on precipices of revolutions, ushering in utopia or the apocalypse – tend to be written by men who have seen Arnold Schwarzenegger’s Terminator failing to assassinate Sarah Connor many times over. If you can also allude to biblical creation, then you’re winning at AI bingo.

AI expert Neil Lawrence springs both traps on page one of his new book, The Atomic Human, and fulfilling the promise of the Terminator’s most quoted line (“I’ll be back”), the film makes a further 15 appearances. Lawrence doesn’t reference the more recent Ex Machina, Alex Garland’s 2014 AI film (that I, full disclosure, had a minor role in creating) which explicitly mocks techbro Silicon Valley arrogance: a Mark Zuckerberg/Elon Musk-like CEO hubristically misquotes a comment made about the creation myth of his own artificially intelligent robot: “If I’ve invented a machine with consciousness, I’m not a man, I’m a God.”

I wonder if we are in a unique moment in history when the discourse is shaped so significantly by the fiction that the men who are in charge of these domains consumed as nerdy youths. Last month it was revealed that Sam Altman, the controversial figure behind OpenAI had asked the actor Scarlett Johansson to voice the latest incarnation of their ChatGPT interface, emulating her role as the AI personal assistant in the 2013 film Her. Johansson declined, but Altman allegedly used a soundalike actor or possibly an artificial Johansson voice anyway. In May, apparently too arrogant (or rich) to bother hiding his deed, Altman tweeted the single word, “her”. Hubris 1: Ethics nil.

We understand big ideas through storytelling. Much has been made of humans as storytelling machines, and Lawrence embraces this mode of science communication with gusto. He indulges us with the Bletchley Park saga, Alan Turing’s brilliance and details of the strategies and technologies of the second world war, in enjoyably exploring ideas of intelligence and how computers can or don’t emulate human cognition. In a chapter called Enlightenment, we veer from Great Man classic tales of Isaac Newton, Winston Churchill and Stephen Hawking, down a cul-de-sac visiting William Blake and Michelangelo, then to Lewis Carroll and Bertrand Russell, and all the way to Elon Musk, via many more.

It says on the press release for The Atomic Human that “the voices of women in AI get overshadowed”. I scanned the index and found that 15 women are named in this 448-page book (16 if you count the goddess Hera), as well as the mention of two groups of anonymous women (Royal Navy Wrens, and the women of Bletchley Park). Winnie-the-Pooh, a fictional bear who as far as I am aware, did not make any pronouncements on intelligence research, or the AI revolution, is mentioned 17 times. I highlight this not to signal my no-doubt jarring political correctness, nor to deny the possibility that, for reasons unexplored in this book, women have played a less significant direct role in the history of AI and big Silicon Valley tech. But if the voices of women are overshadowed in a book by a movie robot or a whimsical bear, then by god we need new stories.

Countless books and films cover the history of Bletchley, of codebreaking, of life in Facebook or Microsoft, about George Mallory and Edmund Hillary, about JFK and the Apollo moon landings. Page after page of the Atomic Human are war stories and rocket stories, jumping about in time and space, and muddling the premise. Maybe I am not manly enough to be excited by this paean to the well-documented but tangential achievements of men. Even if the intended narrative here is to synthesise a thesis about how these well-told tales contribute to our understanding of intelligence, I couldn’t quite pick out the relevance of so many of these boys’ own adventures to the expectation embedded in the subtitle: “Understanding ourselves in the age of AI”.

skip past newsletter promotion

after newsletter promotion

Lawrence is, however, refreshingly dismissive of the tiring amount of posturing and bullshit in the world of AI. I may be in a minority in thinking that the trademark “big concepts” are so often pseudo-philosophical grandstanding by men who like the sport and status but maybe should’ve read a bit more philosophy, and at least had a glance at some history: the singularity – the point when technology is irreversibly beyond our control; transhumanism – a waftily defined state where we are human but vastly enhanced via some unspecified tech; Nick Bostrom’s Superintelligence – a hypothetical technology whose mind far exceeds that of our own crude meat-brains. I’ve never quite resolved whether these future demons are the hooks to get people interested in the real issues, or simply distractions, the magician’s flourish. Either way, I’ve always found them rather tiresome. Lawrence pleasingly labels them as “hooey”, because in all their grandeur, they seem to be terabytes away from the real world of AI that we already live in.

Instead, Lawrence offers novel insight into what intelligence is, how it evolved, and how it is distributed in different living and non-living systems. Comparisons to psychological processing, and the intricacies of the intelligent learning behaviour of our own nervous systems provide insight into the neural processes that do, might or don’t underlie complex artificial administration – for example, in the process of buying something off Amazon, where Lawrence worked for several years – and how much of what is described as AI is merely computation and statistics. This is a salient point that should be better known, that much of the so-called AI in action today is likely to be an Excel spreadsheet doing some numerical powerlifting. But these are points lost in muddled tales whose relevance is often hard to detect.

I would prefer this book half the length and stripped of “great men” of history, to showcase the things we don’t know about what really happens inside the private offices and labs of Amazon or Facebook, and to puncture the movie-baddie hubris. It has an admirable central humanist message: that we are irreplaceable despite the scary waffle of popular discourse. Overall, The Atomic Human is a sensible book, which is higher praise than it sounds, because it tries and to some extent succeeds in rising above the very shallow oceans of public debates about AI that are often shocking but ultimately dull. I just wish he hadn’t started with a tired Uzi-toting cyborg from the 1980s.

NIH-funded research details the brain’s cellular and molecular regulatory elements and their impact on brain function

May 23, 2024
• Press Release

A consortium of researchers has produced the largest and most advanced multidimensional maps of gene regulation networks in the brains of people with and without mental disorders. These maps detail the many regulatory elements that coordinate the brain’s biological pathways and cellular functions. The research, supported the National Institutes of Health (NIH), used postmortem brain tissue from over 2,500 donors to map gene regulation networks across different stages of brain development and multiple brain-related disorders.

“These groundbreaking findings advance our understanding of where, how, and when genetic risk contributes to mental disorders such as schizophrenia, post-traumatic stress disorder, and depression,” said Joshua A. Gordon, M.D., Ph.D., director of NIH’s National Institute of Mental Health (NIMH). “Moreover, the critical resources, shared freely, will help researchers pinpoint genetic variants that are likely to play a causal role in mental illnesses and identify potential molecular targets for new therapeutics.”

The research is published across 15 papers in Science, Science Advances, and Scientific Reports . The papers report findings along several key themes:

• Population-level analyses that link genetic variants, regulatory elements, and different molecular forms of expressed genes to regulatory networks at the cellular level, in both the developing brain and adult brain
• Single-cell-level maps of the prefrontal cortex from individuals diagnosed with mental disorders and neurodevelopmental disorders
• Experimental analyses validating the function of regulatory elements and genetic variants associated with quantitative trait loci (segments of DNA that are linked with observable traits)

The analyses expand on previous findings, exploring multiple cortical and subcortical regions of the human brain. These brain areas play key roles in a range of essential processes, including decision-making, memory, learning, emotion, reward processing, and motor control.

Approximately 2% of the human genome is composed of genes that code for proteins. The remaining 98% includes DNA segments that help regulate the activity of those genes. To better understand how brain structure and function contribute to mental disorders, researchers in the NIMH-funded PsychENCODE Consortium  are using standardized methods and data analysis approaches to build a comprehensive picture of these regulatory elements in the human brain.

In addition to these discoveries, the papers also highlight new methods and tools to help researchers analyze and explore the wealth of data produced by this effort. These resources include a web-based platform offering interactive visualization data from diverse brain cell types in individuals with and without mental disorders, known as PsychSCREEN . Together, these methods and tools provide a comprehensive, integrated data resource for the broader research community.

The papers focus on the second phase of findings from the PsychENCODE Consortium. This effort aims to advance our understanding of how gene regulation impacts brain function and dysfunction.

“These PsychENCODE Consortium findings shed new light on how gene risk maps onto brain function across developmental stages, brain regions, and disorders,” said Jonathan Pevsner, Ph.D., chief of the NIMH Genomics Research Branch. “The work lays a strong foundation for ongoing efforts to characterize regulatory pathways across disorders, elucidate the role of epigenetic mechanisms, and increase the ancestral diversity represented in studies.”

The PsychENCODE papers are presented as a collection on the Science website .

Grants

MH116438 , MH116488 , MH116492 , MH116529 , MH117406 , MH116489 , MH117291 , MH117292 , MH117293 , MH129817 , MH116442 , MH121521 , MH122590 , MH122591 , MH122592 , MH122678 , MH122681 , MH126393 , MH122509 , MH125516 , MH126459 , MH129301 

###

About the National Institute of Mental Health (NIMH): The mission of the
NIMH
 is to transform the understanding and treatment of mental illnesses through basic and clinical research, paving the way for prevention, recovery and cure. For more information, visit the NIMH website.

About the National Institutes of Health (NIH): NIH, the nation’s medical research agency, includes 27 Institutes and Centers and is a component of the U.S. Department of Health and Human Services. NIH is the primary federal agency conducting and supporting basic, clinical, and translational medical research, and is investigating the causes, treatments, and cures for both common and rare diseases. For more information about NIH and its programs, visit the NIH website .

NIH…Turning Discovery Into Health^®