3 ms·
This article provides a lot to think about. To paraphrase some ideas poorly: > "LLMs can make difficult books accessible to more readers (like Cliff's Notes o
by yesfitz 1y ago
This article provides a lot to think about.
To paraphrase some ideas poorly:
> "LLMs can make difficult books accessible to more readers (like Cliff's Notes or Blinkist), BUT some books shouldn't be summarized because the difficulty is part of the importance."
> "An A.I. companion throughout life could be a powerful tool for reflection and memory, BUT my human wife and I (who are capable of love) fill this role for each other."
The author's a reader and writer, and I was initially grateful to see there wasn't pearl-clutching around the impending impact of Large Language Models on literacy. But then I felt like it should have been more of a call to action, either to reject or embrace AI. I wonder if this it the "acceptance" stage of grief for AI skeptics like myself.
- beezlebroxxxxxx 1y ago> BUT some books shouldn't be summarized because the difficulty is part of the importance. It's not even about the difficulty. Why is there this insistent demand to have summaries of everything? Summaries never fully capture a book. If someone told me they only ever read the Cliff's Notes of books, then I simply wouldn't say they read the books. This need to jump to the finish line right away might be more "accessible", but it's also just ignoring and butchering the whole point of the book, which isn't to extract little nuggets of summarizable information, but to be read.