A reader of my newsletter messaged me and asked me this question: How will LLMs change reading and long-form writing?
The following is the answer I gave.
.png)
A Young Girl Reading by Jean-Honoré Fragonard (c. 1770) — Wikipedia
Post type: Stream of consciousness
Epistemic certainty: Reasonably sure
It’s a tricky question, partly because reading and writing are deeply individual pursuits. What they mean to one person can be completely different for someone else.
That said, here’s how I see it.
LLMs are already changing the way people consume content. More people are using them to summarize long-form pieces they might otherwise have read. But it’s also possible those same people wouldn’t have read the full piece in the first place. So it’s hard to say whether the overall effect is negative.
What we do know is that there are early signs of people reading within LLMs. Instead of visiting original websites, they’re consuming summaries and explanations inside the chat interface. This shift is also affecting search traffic. Rather than clicking through to websites, users often get their answers directly in the form of AI-generated blurbs.
Now, will LLMs change how we read and write? I think it depends on what reading and writing mean to the individual.
For many people, reading—especially books—has become a kind of performance. It’s not always about discovery or reflection. It’s about being seen reading. Sharing that you’re reading a certain author. Posting aesthetic photos of your book and coffee cup. Broadcasting a curated intellectual identity.
For me, it’s never really been about that. I used to read in search of wisdom or life-changing insights. But over time, I’ve come to think of books more as both a mirror and a portal. A mirror because they help you notice things in yourself or the world you might have ignored. A portal because they open doors to ideas and places you never would have imagined.
So yes, if reading is just performative, LLMs might completely upend that. But for those who genuinely love the act of reading itself, I don’t think much will change.
Can LLMs change the form of reading? Absolutely. For example, imagine a future Kindle where you can have a live conversation with an LLM about the book you’re reading. That would fundamentally reshape the reading experience. And that’s just one possible direction—there are likely to be many more we can’t even imagine yet.
As for writing, again, it depends on what writing means to you.
Like reading, writing has become performative for many. People write to go viral, to be known, to please an audience. The writing itself often gets reduced to optimization—SEO hacks, clickbait titles, growth tricks.
That’s never been my reason for writing. Of course, it feels good when people read and share what I write. But even if only a few people read it, I’d still keep writing. Writing is how I understand the world. It’s how I think. It’s how I frame ideas and ask better questions. My favorite way to describe it is: reading helps me understand the shape of my own ignorance.
Even with LLMs, my reading habits haven’t fundamentally changed. If anything, I’m reading more books and long-form essays than before. The LLM just acts as a companion. When I read something difficult, I still try to get through it on my own first. But then I’ll use the LLM to clarify, contextualize, or explore related ideas. In that sense, it’s like having a reasonably smart friend in your pocket at all times—someone who knows a lot about a lot.
That’s a powerful thing. But it’s also a trade-off. Using LLMs isn’t free of cost. It changes how we interact with content, and those choices come with consequences.
For people who read or write for external validation—for likes, shares, and recognition—LLMs may fundamentally alter their relationship with both. The dopamine feedback loop weakens when fewer people visit sites or engage with the original content.
But for people like me, not much has changed. I still read. I still write. And I’m using these tools to help me do both, better. I’m also experimenting with AI-assisted writing, co-writing, and more. It’s an exciting space to play in—not a replacement, but a companion.
Learning with Claude
After writing each post, I ask Claude for related ideas so that I can go rabbit hole diving. Here’s what it replied when I shared this post:
Your reflection sits at the intersection of several profound intellectual traditions, revealing how questions about reading and technology echo through centuries of human thought.
The Ancient Roots of Your Concerns: What you’re observing about performative reading isn’t new. Seneca warned against this 2,000 years ago in his Letters: “It is better to frequent the doorways of a few good authors than to wander through the libraries of many.” He distinguished between reading for wisdom (sapientia) versus reading for show (ostentatio). Your insight about books as mirrors connects to Seneca’s belief that philosophy should be a mirror for examining one’s life, not a performance for others.
Media Ecology and the Shape of Thought: Your observations about AI changing reading patterns exemplify what Marshall McLuhan meant by “the medium is the message.” McLuhan predicted that electronic media would create a “global village” but also warned about the “narcosis” effect—how new technologies numb us to their transformative power. Your distinction between genuine and performative reading suggests we’re living through exactly this kind of transition. Media theorist Neil Postman extended this in “Amusing Ourselves to Death,” arguing that the medium shapes not just how we communicate, but how we think.
The Phenomenology of Reading: Your mirror/portal metaphor beautifully captures what phenomenologists call the “hermeneutic circle.” Hans-Georg Gadamer argued in “Truth and Method” that understanding requires a “fusion of horizons”—our present understanding merging with the text’s world. The mirror shows us ourselves; the portal lets us inhabit other worlds. Wolfgang Iser’s reader-response theory describes reading as actively constructing meaning through what he calls “implied readers.”
Writing as Epistemic Practice: Your description of writing as thinking aligns with the rhetorical tradition dating back to Aristotle. But it particularly resonates with Francis Bacon’s revolutionary idea that writing could be a method of scientific discovery. In his Novum Organum, Bacon argued for writing as a tool of systematic inquiry—what we now call the scientific method. Your phrase about “understanding the shape of ignorance” echoes Karl Popper’s insight that science progresses by better understanding what we don’t know—the principle of falsifiability.
Cognitive Symbiosis and Extended Mind: Your AI-as-companion model represents what J.C.R. Licklider envisioned in his seminal 1960 paper “Man-Computer Symbiosis.” Licklider, who later helped create the ARPANET, imagined humans and computers as cooperative partners rather than master-servant relationships. This connects to contemporary work by Edwin Hutchins on distributed cognition—how thinking happens not just in individual minds but across tools, people, and environments.
The Deeper Question: What Is Reading For? Your reflection touches on what Alberto Manguel explores in “A History of Reading”—that reading has never been a stable practice. From oral cultures to manuscript culture to print culture, each transition sparked anxieties about losing something essential. Socrates famously worried that writing would weaken memory in Plato’s Phaedrus. Medieval monks feared that printing would devalue sacred texts. What remains constant is reading as what Paul Ricoeur called a “mimetic” process—not passive consumption but active world-making.
Contemporary Parallels: Your observations align with current research on digital reading comprehension by Maryanne Wolf, who argues in “Reader, Come Home” that digital media are literally rewiring our brains for skimming rather than deep reading. Yet your experience suggests the possibility of what she calls “biliterate” minds—capable of both digital and deep reading modes.
This represents a learning collaboration between human reflection and AI knowledge synthesis, designed to deepen understanding through cross-pollination of ideas.
Timeless Voices on Reading and Books
Some of the most profound reflections on reading capture its transformative power and essential humanity:
“A reader lives a thousand lives before he dies… The man who never reads lives only one.” — George R.R. Martin
“I cannot remember the books I’ve read any more than the meals I have eaten; even so, they have made me.” — Ralph Waldo Emerson
“The man who does not read has no advantage over the man who cannot read.” — Mark Twain
“Books are a uniquely portable magic.” — Stephen King
“Reading is escape, and the opposite of escape; it’s a way to make contact with reality after a day of making things up, and it’s a way of making contact with someone else’s imagination after a day that’s all too real.” — Nora Ephron
“The reading of all good books is like conversation with the finest minds of past centuries.” — René Descartes
“Until I feared I would lose it, I never loved to read. One does not love breathing.” — Harper Lee, To Kill a Mockingbird
“There is no friend as loyal as a book.” — Ernest Hemingway
“Books fall open, you fall in, delighted where you’ve never been.” — David T.W. McCord
“Reading is to the mind what exercise is to the body.” — Joseph Addison
This represents a learning collaboration between human reflection and AI knowledge synthesis, designed to deepen understanding through cross-pollination of ideas.