# How to Summarize Books Using ChatGPT: 7 Experiments in AI Distillation  ## Metadata - Author: [[Tiago Forte]] - Full Title: How to Summarize Books Using ChatGPT: 7 Experiments in AI Distillation - Category: #articles - Summary: insert summary - My notes: - Summary: In "How to Summarize Books Using ChatGPT: 7 Experiments in AI Distillation," the author conducts experiments to test the ability of an AI language model, ChatGPT, to summarize books. One of the books summarized is "Slack: Getting Past Burnout, Busywork, and the Myth of Total Efficiency" by Tom DeMarco. The book challenges the conventional notion that efficiency is the ultimate goal of an organization and advocates for the importance of "slack" in fostering creativity, growth, and adaptability. The author conducts experiments to see how ChatGPT summarizes the book and reflects on the insights uncovered through the process. - URL: https://fortelabs.com/blog/how-to-summarize-books-using-chatgpt/ ## LLM Chats ## NotebookLM ## LLM Audio ## Highlights > Four conclusions from this experiment > First, I concluded that supplying ChatGPT with the best excerpts from one’s reading greatly improves the summary it can provide, especially when those excerpts are organized into an outline that provides a particular point of view. > The main challenge is distilling the excerpts enough to fit within the context window (though upcoming improvements to ChatGPT include a vastly larger window that will fit more than 15 times as much text). > Second, you can’t instruct ChatGPT to come up with “surprising” or “counterintuitive” ways of explaining things, because those qualities depend on sensitivity to a cultural zeitgeist made up of millions of people’s joint expectations, plus access to a wealth of details to put a fine point on its assertions. Instead of *being* surprising, it will merely use words like “surprising” and “counterintuitive” in its otherwise boring description. > Third, ChatGPT is always tending toward cliché, which by definition is the “aggregate average” of the greatest number of writers on which it was trained. You have to constantly come up with inventive ways of getting it off that well-worn path.  > Fourth, I conclude that reading books is still worthwhile, even in an age of Artificial Intelligence. Besides the personal enrichment from the experience of reading, there are key details, subtle distinctions, and gestalts of meaning that come from reading and notetaking that can’t (yet) be reproduced by even the latest language models.  > In fact, it seems that having a personal collection of excerpts from one’s reading, whether written down by hand or synchronized with a notetaking app using a service like [Readwise](https://readwise.io/basb/), is an even greater asset than before. The decisions you make about which excerpts are worth saving constitute a unique interpretation of a book’s ideas that isn’t shared by anyone else on the planet. > **In a world of infinite, instantaneous summarization, the one with the most unique data to summarize wins.** ([View Highlight](https://read.readwise.io/read/01hk2mnp94wt0he6amqtgegg6z)) - Tags: [[Chat about aesthetics of ai 9 Sep 2025]]