IO reader
New way to read. Read deeper. Think Sharper.
11 followers
New way to read. Read deeper. Think Sharper.
11 followers
A reading environment that trains deep focus: the great books, immersive reading modes, and an AI that challenges you to think — not one that thinks for you.
How does the AI actually challenge you without just giving answers away — does it ask Socratic questions or flag passages you might be glossing over?
@sedanurnnjz Both, depending on which persona you tap. The Questioner runs Socratic method — it answers your question with the question underneath it. The Critical Thinker is the flagging one: it names the assumption in the passage you were about to swallow and hands it back to you. The one rule shared by all 12 personas is that none of them may give you the conclusion — no summaries, no "what this passage means." I walked through a concrete Dostoevsky exchange in my reply to Erdem above if you want to see the actual shape of it. I suggest you to download the app and give it a try. 60 AI responses are free!
Curious how the AI actually pushes back without just summarizing or giving away the answer. What does a typical challenge look like while reading something dense like Kant or Dostoevsky?
@erdemzdzmxx3 Good question — easiest to just show one. Here's the passage in Crime and Punishment where Raskolnikov lays out his "extraordinary man" theory (Part 3, Ch. 5):
"...an extraordinary man has the right... an inner right to decide in his own conscience to overstep certain obstacles, and only in case it is essential for the practical fulfilment of his idea (sometimes, perhaps, of benefit to the whole of humanity)."
A summary product tells you: "Raskolnikov argues great men may transgress moral law for humanity's benefit."
True — and now you've skipped the most argued-about pages in the book.
Tap that paragraph with the Critical Thinker and this is what comes back — actual output, unedited:
"Raskolnikov's elaborate justification for 'extraordinary' men to transgress the law hinges on the retrospective validation of their actions by history, implying that success alone legitimizes their methods. This perspective conveniently sidesteps the immediate, tangible suffering inflicted by such transgressions, reducing victims to mere obstacles in the path of progress. The argument also fails to provide a clear, objective metric for identifying these 'extraordinary' individuals before their actions are judged by history, creating a dangerous loophole where anyone could claim this 'inner right' to violence. How does one distinguish a true 'man of the future' from a mere psychopath before the historical verdict is rendered?"
No verdict on the book, no summary — an objection you have to take back into the text to answer. (Raskolnikov's examples are Newton, Kepler, Lycurgus, Napoleon — all men history already vindicated; the AI is pointing out he needs the label before the act.) That's the design test for all 12 personas: if you can respond to the AI without rereading the passage, the persona failed.
Kant gets the same treatment. The Groundwork's shopkeeper passage is a favorite, because the honest-from-prudence vs. honest-from-duty distinction is asserted more than argued there, and the persona will point that out and make you defend it.
One honest caveat: it's a language model, so quality tracks the passage. On argument-dense text like this it's consistently sharp; on pure plot pages it has less to bite into.
the immersive reading modes look beautifully restrained, like they actually thought about how a page should feel rather than just piling on features
@elifsuowp1 That's the design contract in one line — the question for every screen was what a page needs, not what an app usually has. Square geometry, no shadows, print typography, nothing that moves unless you moved it. Glad it reads that way.
I like that you're treating AI as something that challenges the reader instead of replacing the act of thinking.
As models get better at generating answers, I think products that strengthen human reasoning become more valuable than products that simply produce more text.
Helping people think better feels like a durable direction.
@aryan787544 Happy to meet someone who is sharing the idea about value of AI and how to adopt it to the product. :) Please try to use IO reader and share your feedback. I think there will be areas to improve further and your feedback will be valuable.
@min_byun I'd be happy to.
One thing I'll be paying close attention to isn't just whether the AI helps people think more deeply, but whether it changes how they approach reading after they close the app. That's a much harder outcome to create, but it also feels like the difference between a useful AI feature and a lasting reading habit.
I'll spend some time with IO Reader and share my thoughts.
@aryan787544 That sounds great! Please rate and write review in Apple App Store https://apps.apple.com/us/app/io-reader-read-deeper-with-ai/id6767890380 . That will be greatly helpful.
Genuinely impressed by how the AI nudges you to wrestle with ideas instead of spitting out answers. The reading mode is clean and keeps me from drifting off.
@tolgakaraat "The reading mode is clean and keeps me from drifting off." is exactly what I wanted to achieve through this app. Thanks a lot for your valuable comment and please let me know if there is anything to further improve.