This article is republished with permission from Wonder Tools, a newsletter that helps you discover the most useful sites and apps. Subscribe here.
Imagine turning your reading history into a treasure map. By feeding a list of your favorite books and movies to an AI assistant, you can uncover hidden patterns in what you love. From your subconscious attraction to unreliable narrators to your love for stories that begin at the end, you may be surprised by what an AI assistant can reveal.
Building a personal taste atlas helps you understand your reading self better. It can also surface blind spots in your cultural diet and point you toward unexplored literary territories youre likely to love.
Why analyze your preferences?
This isnt just another recommendation engine. Netflix or Amazon may suggest what to watch or buy next based on viewing history, but a taste atlas goes much deeper.
It analyzes themes, narrative structures, and emotional resonance across media formats. It can reveal connections between novels you adore and foreign films youve never heard of, or help you articulate why certain stories stick with you while others dont.
You can tune the atlas by adjusting the info and examples you give it. You can customize the analysis with your prompts, asking for particular kinds of observations or recommendations.
With AIs help, you can map out your own universe of awesome. As you scout out gaps in your reading or movie watching, you can discover authors and films that expand your horizons.
Start by gathering your favorites
You need to provide an AI assistant with a list of at least 10-15 titles that resonate with you for meaningful insights; 30+ is better. Here are the fastest ways to gather them.
Physical books or DVDs: snap a photo of your bookshelf. AI can read the titles. Or write a list of titles on paper. AI assistants can read handwriting surprisingly well.
Digital readers: refer to your Kindle library, your read shelf on Goodreads, listen history on Audible, timeline on Libby, or any doc or spreadsheet you maintain with your favorites.
Streaming: Apps like Likewise, Sofa, Listy, Listium, Letterboxd, Trakt, and Reelgood let you compile lists of favorites. You can use those collections to train your AI assistant.
Use your voice: If talking jogs your memory, use conversation mode in ChatGPT, Claude, Googles Gemini, or Microsofts CoPilot. Let the AI interview you about your favorite books or movies.
Scan award lists: If you cant think of favorites, check a list of Oscar-winning movies or book awards for reminders of what youve enjoyed.
Criteria: Consider titles you often revisit or recommend. Include recent favorites and older resonant ones. Give extra weight to those that provoked emotion, changed your perspective, or prompted action. Ideally, note not just the title but one or more aspects of a work that particularly resonated.
Prompt AI to analyze your list
Once you’ve compiled your list, use your preferred AI tool to uncover patterns in your literary tastes. Prompt the AI assistant for insights to advance your self-understanding. After that, ask it to help you discover more books/movies you’ll love.
Start by writing a detailed prompt to elicit a thorough, subtle analysis of your taste in books or movies. Heres an example you can adopt or adapt:
You are a perceptive literary critic and cultural analyst with deep knowledge of literature across genres and cultures. Carefully analyze the attached list of my favorite books for patterns. Think deeply about connections between titles and topics that might not be immediately apparent. Where you notice interesting patterns, explain your reasoning and cite specific examples. Please analyze this list of my favorite books. Create a detailed literary taste profile that identifies:
Core Elements:
Primary themes and topics
Genre preferences and style patterns
Narrative approaches and structure choices
Character types and relationships
Tone and emotional range
Upload a file with your list or paste it.
Which AI tool to use?
ChatGPT 4o worked well for me in importing Google Docs and PDFs with my favorites. Its analysis and recommendations were nuanced and helpful.
Limitation: Occasionally, it suggested authors who were already in my existing lists, despite being prompted not to.
Claude Pro provided an excellent overview of the kinds of books Ive selected for the book group I facilitate over the past eight years. It helped identify gaps in our reading list and offered useful suggestions for future titles.
Limitation: Some documents I tried to import, like my Readwise reading highlights, were too large to fit in a Claude Project I created for my taste atlas.
Gemini 2.0 Experimental Advanced, Googles newest model, was an excellent voice partner in analyzing my current reading interests.
Limitation: 2.0 couldnt yet import documents, but Gemini 1.5 could. It helpfully analyzed the Google Doc with my complete Readwise Highlights archive.
Use either free or premium AI tools for this analysis. For long book lists or extensive highlights, use a pro model for nuanced analysis.
Expand your taste horizons
Once an AI tool has analyzed your book or movie preferences, prompt it to suggest new authors and titles. Ask about specific connections between the titles you liked and its recommendations, so you understand the rationale.
Cultural leaps: Ask AI to identify authors who write like your favorites but in different languages or cultures.
Whats missing? Try a prompt about negative space what authors, titles, topics or genres are missing from your favorites. What notable titles might stretch your literary horizons?
