[_bramses]

Hopefully in the future, language models will be able to read books for us and summarize whether or not they’re worthwhile. Bumping from 200 wpm to 300 wpm doesn’t put a dent in the 100+ million books circulating in print.

Unrelated Readwise questions (love the work y’all do, thank you!) - any idea when the chrome app will support multiple highlights in one go? - is there a random quote API that references all sources at once? - has the team considered adding semantic search to the app?

[prometheus76]

We already have rating systems and reviews that are human-generated and human-curated to help you decide if a book is "worthwhile". How could an AI language model improve on that? In other words, why would you trust AI more than human curators/reviewers?

[kadoban]

An AI could tailor it to the person, put it in context of other things already read, and essentially create a course of study for you. Humans can do the ~same, but it's pretty expensive.

An AI could also improve on the status quo by being more consistent and there's _some_ possibility that it could be less biased than in human in several ways.

[hoodwink]

hey there! thanks for the kind words :)

on AI-based booked summarization, here's a really interesting article by OpenAI on the topic: https://openai.com/blog/summarizing-books/

*any idea when the chrome app will support multiple highlights in one go?* we have a new browser extension (yet unreleased) that enables you to highlight the native page (i think that's what you're asking)

*is there a random quote API that references all sources at once?* at the individual user level, yes. readwise.io/api_deets

*has the team considered adding semantic search to the app?* not yet, but my cofounder tristan did push a huge full-text search update a few months ago that makes search results 10x better than they were before

[layer8]

Different people will have different opinions on which parts and aspects of a book are worthwhile or not. Books that were worthwhile to me ten years ago may not be worthwhile to me today, or in ten years, and vice versa. I don’t expect AI to be able to adequately address that anytime soon.

[geysersam]

Sounds like a book review.