Reading Time: 4 minutes

One of the advantages of reviewing books over a long period of time is that you eventually build a record of your interests. The challenge is that after dozens of reviews spread across years, it can be difficult to step back and see the bigger picture.

Recently, I decided to do exactly that.

Not by manually sorting through every review, but by using artificial intelligence as a research assistant.

The result was surprisingly insightful.

What began as a simple question—”Based on the books I’ve reviewed, what books should I read next?”—turned into an interesting exercise in understanding my own reading habits.

The Experiment

I asked AI to look at the books I had reviewed and discussed over time and identify patterns.

The first step was compiling a list of books I had already read.

Among them were:

• Grant by Ron Chernow
• Churchill: Walking with Destiny by Andrew Roberts
• The Soul of America by Jon Meacham
• And There Was Light by Jon Meacham
• The War on Peace by Ronan Farrow
• Gulag: A History by Anne Applebaum
• Say Nothing by Patrick Radden Keefe
• The Warmth of Other Suns by Isabel Wilkerson
• Caste by Isabel Wilkerson
• Several novels by Barbara Kingsolver
• Multiple historical novels by Ken Follett
• The Poldark series by Winston Graham

After compiling the list, I asked a simple question:

“What do these books have in common?”

The answer was more revealing than I expected.

AI Found Patterns I Hadn’t Fully Recognized

At first glance, the books appear to be all over the map.

Biography.

History.

Politics.

Historical fiction.

Social commentary.

Current events.

But when viewed collectively, clear themes emerged.

The AI pointed out that I consistently gravitate toward books that explore:

• Leadership
• Human behavior
• Institutions
• Society
• Political systems
• Historical change
• Cultural conflict

In short, I seem to be less interested in isolated events and more interested in understanding why people and societies behave the way they do.

That observation felt accurate.

I Am Less Interested in Facts Than Context

I enjoy history, but I rarely read history for dates and timelines alone.

What interests me most is the human side of the story.

How do leaders make decisions?

Why do institutions succeed or fail?

What causes societies to change?

How do ordinary people respond when faced with extraordinary circumstances?

Whether the subject is Abraham Lincoln, Ulysses Grant, Winston Churchill, or a family living through social upheaval, I find myself drawn to the people behind the events.

That was one of the strongest themes AI identified across my reviews.

Historical Fiction Serves a Purpose Too

Another observation involved my fiction reading.

I’ve spent a fair amount of time reading authors such as Ken Follett and Winston Graham.

Initially, I thought of these books as a separate category.

The AI saw something different.

It recognized that many of these novels explore the same themes that attract me to nonfiction:

• Leadership
• Conflict
• Social change
• Human resilience
• Historical forces

The format is different, but the underlying interests are remarkably similar.

Apparently, I read for insight, regardless of whether the book is fiction or nonfiction.

Turning Reading History Into Future Recommendations

Once the patterns became clear, I asked the next logical question:

“Based on these interests, what should I read next?”

Instead of recommending random bestsellers, AI-generated suggestions aligned with my established interests.

Some of the strongest recommendations included:

History and Biography

• Master of the Senate by Robert Caro
• Destiny of the Republic by Candice Millard
• The Great Influenza by John M. Barry
• The River of Doubt by Candice Millard

Society and Culture

• The Righteous Mind by Jonathan Haidt
• American Nations by Colin Woodard
• The Worst Hard Time by Timothy Egan

Historical Fiction

• The Winds of War by Herman Wouk
• War and Remembrance by Herman Wouk
• The Source by James A. Michener

What I appreciated was that these recommendations were not based on popularity.

They were based on patterns found within my own reading history. They are books I would like but probably never considered or found.

AI as a Reading Assistant

This experience reminded me that AI can be useful for much more than researching and correcting article mistakes.

In this case, it acted more like a librarian, researcher, and reading advisor.

It helped:

• Compile books I had discussed or reviewed
• Identify recurring themes
• Categorize interests
• Highlight connections between books
• Recommend future reading based on actual habits rather than trends

The recommendations weren’t random.

They were personalized based on years of reading choices.

What Surprised Me Most

The biggest surprise wasn’t the list of recommended books.

It was when I realized that my reading habits have been remarkably consistent.

I thought I was reading a wide variety of topics.

In reality, I have been exploring many of the same questions from different angles for years.

Questions about:

• Leadership
• Human nature
• Society
• Power
• Institutions
• History

The books changed.

The questions remained the same.

Final Thoughts

When I started reviewing books, I assumed I was creating a record of what I had read.

What I was actually creating was a record of my curiosity.

Using AI to analyze that history helped me see patterns that would have been easy to miss if I looked at each review individually.

It also gave me a roadmap for future reading.

Perhaps the most interesting lesson was that our bookshelves often tell a story about us.

Not just where we’ve been, but what we’re trying to understand.

In my case, AI helped connect the dots.

The result wasn’t simply a list of book recommendations.

It was a better understanding of why certain books continue to resonate with me year after year.