Why do people keep writing, singing, and building for audiences the market can’t see? The answer to this question is what makes some people irreplaceable.

I just overheard a conversation between two strangers that belongs to our cultural moment: they were two grown male teenagers, and one asked the other whether he had read a book I didn’t recognize.

The other, half-mockingly, half-seriously, asked rhetorically, “Who reads books all the time anymore?” I thought he might have a point. I thought about how guilty I sometimes feel for reading for long periods instead of “engaging” with work and reality more directly (as if reading long-form were indulging).

Era of summaries

It’s not new that some people aren’t reading, or that competition from other media that offer higher dopamine rewards with lower cognitive effort is diminishing the appeal of serious books among occasional readers. What’s new is that this fact isn’t a cause for concern and has become almost part of the bragging rights of those who claim to be too busy to read or even listen to entire books.

There are many reasons why books are more often summarized and discussed on live streams and podcasts than read individually, for the sake of it. It’s a trend that clearly predates chatbots.

That said, AI tools are changing the way we conceptualize ideas and reality; simply asking for summaries of everything, secretly assuming that the mechanical summary of a complex text, or the automation of a task, is close enough to pass for the thing itself. I can understand that a haiku, an aphorism, an epigraph, an anagram, etc., can contain a world inside, and they are as short as you can get by design. But let’s not kid ourselves into thinking that a dry summary will give us what a book can.

The deferred library

I remember once, when talking to an acquaintance (a very intelligent, successful man in San Francisco), mentioning a book in the context of our conversation. He shrugged off the reference, virtue-signaling that, since he was so busy, he hadn’t read books regularly since college; in his words, the task required too much time and attention, and he could allocate that capital to his job. Then, after professional life and economic safety (in his world, a concept also known as “fuck you money”), he’d be able to read as much as he could.

Who is reading long form? And how many people, reaching adulthood, decide to cultivate the habit of reading fiction, recommended and at random, regularly and without guilt? I figured that many things have been changing for a sizeable number of people, and perhaps we’re culturally relegating reading fiction to unproductive, indulgent leisure. Is that so, and how generalized is the phenomenon? Is it happening on both sides of the Atlantic, and if so, why?

Before reaching any conclusions too quickly, I’ve also been following the latest studies and articles dedicated to a generalized shift online from web surfing (exploration of links offered by a conceptual query on a search engine) into asking for AI summaries, both in search engines (which are turning into AI tools themselves) and chatbots.

Twelve open tabs

Not long ago, if we wanted to get a sense of where to start when trying to read a book or what a movie from a certain writer or filmmaker, we performed a search, then visited different types of information to reach our own conclusion: several reviews by traditional media, some popular results by random social media users and so-called influencers, a couple of relevant threads on forums like Reddit, perhaps a listicle or two…

After visiting topics of interest, sometimes stored across different website tabs, we’d follow a lead, often by intuition, and that gut feeling would prompt action, either immediate (reaching a long-form document online, a movie available on streaming, etc.) or not. It’s the same gut feeling we call personal taste when we visit a place and enter a store at random, almost always finding some connection to the artifacts displayed, whether it’s books, clothes, food, art, building materials and finishes, etc.

There’s nothing wrong with summaries and AI, which are already changing the way we work and use the internet. But as we increasingly rely on chatbot responses, what happens if people publishing content on the internet don’t find a way to break through summaries and, as visits to sites decrease, most articles and content become worthless to people creating them?

What the trackers saw

A Pew Research study from a year ago already showed the trend. Drawing on browsing data from 900 U.S. adults, the study showed that users clicked a link in just 8% of searches that displayed an AI summary; interestingly, in searches not showing an AI summary on the topic, they clicked 15% of the time. Searches with summaries also led to more users abandoning browsing on the topic altogether after reading the summary (26%), compared with 16% on pages with no contextual AI enabled.

Visits to sources on desktop computers (CTR, or “click-through rate”) have dropped by up to 50%, according to Search Engine Land/Kevin Indig, and Authoritas reported a drop of up to 79%. The study was conducted by installing a tracker in the participants’ browsers, so it observed behavior rather than reporting it, which has led people to conclude that online search is morphing into a different tool, with AI summarization taking over web surfing.

So, we are beginning to access information, both long-form and contextual internet activity, in a different way: AI overviews and summaries showed up on 6-7% of searches at the start of 2025, whereas by February 2026, the same summaries (often devoid of visible links to follow through) existed on 48% of queries, up 58% year on year, according to BrightEdge.

Given how rapidly our behavior has changed, let’s consider an extreme case scenario. If summaries get better and better, and AI companies manage to circumvent the copyright guardrails that seem to be falling once again regarding human creation like books, music, or movies, perhaps the next content frontier attempted by corporations providing mass AI tools would be a new type of “abridged” versions of celebrated units of human knowledge like scientific studies, non-fiction and fiction books, and any other sort of digitized masterpieces.

Pushed to the extreme, overall we would then become a civilization relying entirely on summaries (long and short) and derivatives of previous human work, disincentivizing the ancient human pursuit of intellectual creation as an activity through pay-per-use and/or ownership.

The veteran with the ruined left hand

Of course, there are many incentives beyond economics, and many people create things, digitally or otherwise, because they feel the urge or call to do so despite lacking a direct reward or a sizeable audience paying attention. Now, with the internet, and also before the rise of our current digital lifestyles, only a minority of creators lived comfortably off their work, and only a select bunch reached universal recognition and riches within their lifetime: the likes of Shakespeare and Cervantes, Molière and Lope de Vega were celebrated in life and lived by their pens — Shakespeare, the shrewdest businessman among them, rather better than that — but nothing like the industrial fortunes to come.

Later on, some authors were modern best-sellers and became the most influential people of their time: Walter Scott, Charles Dickens, Alexandre Dumas, Jules Verne, Victor Hugo (perhaps the first “engaged intellectual,” influencing politics and society with his work)…

Cervantes famously wrote what’s arguably the first intended modern novel (something that would attract both cultured people and the common folk, getting to the work through read-alouds, like the picaresque genre or satirical poetry created to attack notable people of the time covertly) so he could sell enough copies to settle his accounts with the Treasury and plan a decent retirement as a maimed war veteran. More than glory, the veteran with the ruined left hand — himself a former tax collector, jailed when his accounts failed to balance — needed to eat and shake off the Crown’s auditors.

Like many other authors and creators, Cervantes was far from being someone detached and inconsequential, and he was described as a brave infantry soldier aboard one of the galleys, the Marquesa, representing Spain under John of Austria at the Battle of Lepanto against the Ottomans on October 7, 1571. The creator of Don Quixote was perhaps a mess of a citizen, but nobody can say he didn’t show any urge to pursue adventures in life, as in literature.

The overflow

Many authors have reflected on the profound incentives of human creation, and there’s much more to it than the pursuit of glory or mere economic need. Nietzsche’s response to the same question (“why create?”) evolved throughout his lifetime. During his youth, when he considered composer and friend/mentor Richard Wagner the most important creator of the time, Nietzsche thought that many people needed to transcend with art because their aspiration was what made an imperfect and cruel world inhabitable; later, already at odds with Wagner and convalescent, he saw creation as the affirmation of one’s existence.

From this lens, the creator doesn’t need to succeed (either economically or with the public) to reach a higher level of self-realization. According to Nietzsche’s life-affirmation hypothesis, some people need to express their overflowing energy through an act of creation that becomes an underlying life purpose and gives meaning to it all, even if it doesn’t earn public approval or pay the bills. The reductionist causality between creation and material rewards didn’t foster the first epic poems, foundational to every major culture, all of which emerged from oral tradition.

To the question “why create?”, many, like Nietzsche, think it’s a human need: there are people who need to occupy themselves with intellectually engaging endeavors as overflow of their energy and personality, irrespective of whether they’ll find an audience, let alone applause and economic reward.

An audience of One

The addressed audience is sometimes so small that it consists of one other person besides the creator, and sometimes it’s only the creator himself. When he took over the task of building the grandiose (and controversial from the beginning, given the scale and radical style) Sagrada Família, Catalan architect Antoni Gaudí expressed his determination in a way that might have scared the naysayers: his “my client is not in a hurry” showed that, to the architect, the work was above any worldly reward.

He’s seldom the only one, and we’ve certainly lost the opportunity to enjoy works and masterpieces due to the relative obscurity or zeal of some creators. Consider, for example, Gerard Manley Hopkins, an Oxford aesthete from the nineteenth century; as G.K. Chesterton would do decades later, Hopkins converted to Catholicism, but, unlike Chesterton (who sought out celebrity), Hopkins, a gifted poet, had an audience of One (or two, himself and a superior being, just like Gaudí).

When he converted, he studied to become a Jesuit and, before entering the novitiate, he burned his poems without publishing them (which he called the “slaughter of the innocents”). Versions of his poems survived in notebooks and with friends. Years passed, and when, in 1875, a ship wrecked in the Thames estuary, drowning Franciscan nuns escaping from persecution in Germany, he was asked to write a poem by his rector. But the poem was so radically different, so metrically advanced for his time, that the zealots controlling The Month, a publication by his order, rejected it.

That’s when he stopped submitting, but that didn’t deter a select group (Robert Bridges, Canon Richard Watson Dixon, Coventry Patmore) from enjoying the little work they knew of Hopkins. His story is probably the closest thing in the nineteenth century we can think of to cult musicians, writers, or filmmakers whose quality and authenticity never guaranteed an audience for whatever reason.

Who’s your Socrates?

Argentinian writer, encyclopedic reader, and Anglophile Jorge Luis Borges, who never wrote a big novel nor was considered a philosopher, a semiologist, or a mathematician but contributed more to those fields than most people, was a legendary conversationalist when he found the right mind to talk to.

When meeting his friend and fellow Argentinian writer Adolfo Bioy Casares, Borges would recall many of his literary adventures (the paths taken, and also those adventures and challenges not taken), and they certainly talked about the life and character of an author that they considered their literary master, a profound philosopher, and a personal mentor to both of them, poet Macedonio Fernández.

Borges explained once and again that Macedonio seemed to recite all the poems worth knowing in the world, and he also created the most beautiful ones, most of which he had annotated in a hurry on a piece of paper somewhere, kept in the pocket for a couple of days, and disposed of.

To their generation, old Macedonio Fernández was their Socrates: he cared more about real-time thinking and the purest art of creation than publishing. Borges confessed that he had imitated his master for years, to the point of “impassioned plagiarism,” yet he knew that, with the years, he would be published and remembered, and so would Bioy Casares, while their master — who published little in life, and that reluctantly, at his friends’ insistence — would survive mainly as a rumor transmitted by the people who had heard him think. Fame or posterity weren’t his drive.

The harp string

When I think of people like Gerard Manley Hopkins or Macedonio Fernández, I realize that the human need for creation, exploring truth and beauty with their own expression as an act of affirmation, can’t be fully disrupted by any process of mechanical knowledge aggregation and refactorization such as AI, even when these models keep improving and surprising us in many fields.

I presume that AI will never be capable of trembling with the intuition of beauty, purity, and personal truth as many human creators do, and that’s probably what will save us. Nietzsche, from The Will to Power (trans. Walter Kaufmann and R. J. Hollingdale, Vintage, 1968):

“If we affirm one moment, we thus affirm not only ourselves but all existence. For nothing is self-sufficient, neither in us ourselves nor in things; and if our soul has trembled with happiness and sounded like a harp string just once, all eternity was needed to produce this one event — and in this single moment of affirmation all eternity was called good, redeemed, justified, and affirmed.”

The story of Sixto Rodriguez is almost too unreal. A mysterious man living in the decomposing working-class underworld of Detroit in the late sixties and early seventies, Rodriguez combined odd jobs in construction and hauling with small singing gigs with his guitar in ill-reputed joints, where his talent for writing and unique voice were discovered by two producers out of Motown’s session world, Dennis Coffey and Mike Theodore, who helped him produce two records that went absolutely unnoticed in the United States.

Dead on stage, alive in Detroit

Somehow, one record made it to South Africa, whose population was trying to cope with being considered an international pariah at the apex of apartheid. Back then, records were censored in South Africa, yet many of Rodriguez’s songs were passed around, becoming a mass phenomenon.

Rodriguez’s debut album, Cold Fact, was so popular and so widely shared in South Africa that the bootleg success would reverberate into the future, but at the time nothing was known about the author, and word spread that he had ended his life on stage, setting himself on fire, whereas other articles and fanzines claimed it had been a bullet to the head.

The reality was wilder than the rumors: the mysterious Rodriguez, of profound voice and mystical, sorrowful lyrics, the son of Mexican immigrants, was still working for little and living in an unheated, dilapidated Detroit house. Rodriguez never stopped playing music and composing songs; only, he did it at home, alone with his guitar next to the window, as it rained or snowed outside — apart from two Australian tours at the turn of the eighties, a detail the legend, and later the film, preferred to forget.

But almost thirty years later, when a South African crew of fans and indie journalists contacted his family, he probably realized the songs weren’t meant to stay only with him. Searching for Sugar Man, a 2012 documentary film by Malik Bendjelloul, can tell us so much about the parts that machines won’t take from people who seek expression and creation for its own sake, rain or shine, to an audience of one or many. To them, it matters little.

Footnotes to an invisible text

Human creation doesn’t depend on remuneration and utility as much as incentives-obsessed economists think. There are many Macedonio Fernández and Sixto Rodriguez types in this world, and many of us are lucky enough to know or have known incredibly talented people whose drive in life isn’t status, so they can dedicate more time to what matters to them.

A writer from Barcelona, Enrique Vila-Matas, imagined a middle-aged office clerk who had published one novel twenty-five years earlier, going mute right afterward. He breaks his silence by writing a diary composed of footnotes referencing a text that doesn’t exist, which, according to the clerk’s vision, is a representative catalog of some of the most authentic writers who, for one reason or another, stalled in their careers and stopped writing altogether.

Vila-Matas’ novel, Bartleby & Co., is named after Melville’s short story Bartleby, the Scrivener (a recently hired Wall Street clerk shows an initial burst of hard, competent work, but then refuses to do any task, responding “I would prefer not to.”) The book’s epigraph is from Jean de La Bruyère: “The glory or the merit of certain men consists in writing well; that of others consists in not writing.”

The book’s initial few pages are hilarious, and they can also remind us that, sometimes, the song we’d like to listen to or the book we’d like to read was lost to time or was never written, but the mere potential of it existing in someone’s mind and aspirations is what makes us impossible to be replaced by machines. After the opening pages, the rest of the novel is a series of footnotes. The meta-text becomes the entire work.

Mourning versus forgetting

Of course, Vila-Matas (or the clerk who is the narrator and writer of 86 footnotes to a text that doesn’t exist) was onto something when writing this novel that discussed the crisis of the novel as a genre, as it has to compete with other entertainment formats.

A book made entirely of annotations to a missing text is the dignified mirror of the digital world that we are building, which is made of summaries of sources that nobody visits.

However, there’s one big difference between the thesis of Bartleby & Co. and the race to summarize the world’s knowledge and creation: the clerk writes the footnotes mourning the text that should have existed, understanding the emptiness left by its absence. What we are doing now with summaries is consciously forgetting there was one text, because it takes too much effort to go through it and enjoy it without guilt.

To the question that opens this article (who reads books all the time anymore?), I’d say: those who suspect that a summary of a book is not a book, any more than a summary of a life is a life.