The Age of Reading Is Over. But Is It?

The invoice

The numbers are real.

It is worth taking them seriously rather than waving them away as nostalgia. Federal survey data show the share of U.S. adults who read any book in the prior year falling from about 55% in 2012 to under half by 2022 — the lowest fiction-reading rate in the survey’s thirty-plus-year history.¹ Across two decades of a time-use survey covering hundreds of thousands of Americans, the share reading for pleasure on a given day roughly halved, from around 28% to 16%, with the steepest drops among lower-income and less-educated readers.² The nation’s own report card tells the same story from the other end of life: record shares of students reading below the “Basic” level, and the lowest-ever twelfth-grade reading score.³

This isn’t illiteracy. It’s post-literacy: we are still surrounded by words, but losing the specific, effortful habit of holding a long thought in the head long enough to be changed by it.

The cause

It wasn’t the phone. It was the software.

The comfortable story blames phones, and there is truth in it. But blaming the hardware lets the software off the hook, and the software is the part we actually chose to build. For fifteen years the most capable engineering talent alive has been pointed at a single objective: remove friction. Make it faster. Make it easier. Require less of the person.

The feed removed the friction of deciding what to look at. The summary removed the friction of reading the thing. The answer engine removed the friction of not knowing. Each of these felt like a gift. Each quietly removed a rep from a muscle.

The muscle

And muscles atrophy.

Reading is that muscle. Deep reading — the kind where you lose the room — is not natural. It is a trained, fragile skill sitting on top of a brain evolved to scan the savanna, and the researchers who study it have shown the “deep-reading circuit” is built rather than given, and can weaken under screen-driven skimming.⁷ A skill built by effort can be un-built by convenience. You don’t lose it the way you lose your keys. You lose it the way you lose a language you stopped speaking.

And the effort was never the tax on understanding. Half a century of learning science points the other way: with the right kind of book, the difficulty is the mechanism of learning, not an obstacle to it — the very convenience of a shortcut is often the signal that little is being learned.⁶

The uncomfortable part

I built an AI app.

In an essay about the collapse of reading, that should set off an alarm — and it should. The most on-trend product I could have shipped this year is an AI that reads books for you: point it at War and Peace, get four bullet points, feel briefly educated, move on. There are a lot of those now, and the early evidence on the trade they offer is not reassuring. In a study of hundreds of people, heavier reliance on AI tools tracked with weaker critical-thinking scores, the link running through “cognitive offloading” — letting the tool do the thinking.⁴ An MIT experiment wired up essay-writers and found those leaning on a chatbot showed the weakest brain connectivity and, afterward, mostly could not quote a single line they had just “written” — what the researchers call cognitive debt.⁵ Both come with honest caveats — one correlational, one a small preprint — but they point the same way: when the machine does the work, the mind that should have done it comes away with nothing.

So let me be plain about the obvious objection, because someone will raise it anyway: an AI company writing against AI is a contradiction. It would be, if that were the argument. It isn’t. The argument is against AI that replaces reading — and I have never hidden that IO is AI, only pointed it the other way.

What I built instead

An AI that refuses to summarize.

IO reader’s twelve personas are not there to hand you a takeaway. They are there to help you read the actual page more deeply. A Historian places a passage in the moment it was written. A Rhetorician shows you what a sentence is doing to you — the word choice, the rhythm, the move you felt but couldn’t name. A Critical Thinker surfaces the assumption buried in the text that you were about to read straight past. A Questioner does the one genuinely difficult thing: it asks you escalating questions and makes you find out whether you actually understood.

None of them hand you a conclusion. A summary replaces the book; these surround it. You still read every word. That is not a limitation I apologize for — it is the design. I made the fuller case for it elsewhere, and the whole evidence base — the studies behind every claim above — lives in our research review.

I know how this sounds. The phone is where attention goes to die, and here I am putting a reader on it — the people who take reading most seriously reach for e-ink precisely to get off the glass. But most people won’t, and I didn’t want to build only for the ones who already read. So instead of pretending to be paper, IO does something less precious: it borrows the design that makes the feeds so hard to put down — the scroll, the one-thing-at-a-time swipe — and points it at the canon. Reading in it can be a feed where every post is a paragraph of the book, or a swipe where each card is a single line worth sitting with. The difference from the feed you would otherwise be in isn’t the interface — it’s that you chose the book, nothing is deciding your next hit, and the AI is there to deepen the page rather than summarize it away. It can’t stop you leaving for TikTok. It is built to be the thing you would rather not leave.

Easier was always the disease. The effort was the reading.

So the question I kept asking while building this was never “how do I make reading easier.” It was whether you could point the same technology at the opposite goal: make a hard book reachable, and richer, without doing the reading for you. An AI that reads with you, the way a good professor or a well-read friend does, rather than one that reads instead of you.

It is one person’s bet, live on the App Store, and honestly almost nobody has noticed it yet. But I am increasingly sure the bet is right, because the whole industry is leaning the other way and The Atlantic’s cover is the invoice for that lean. The age of reading is not over. It is being ended, by software built to spare us the effort — and that means it is a choice, which means it can be chosen differently. The most valuable thing software can do now is not save you another minute. It is give you back the capacity for the harder thought, and be honest that the effort was never the bug. It was the reading.