AI isn’t for summarizing books. It’s for arguing with them.

You shouldn’t use AI to read faster.

You should use it to read deeper.

Because the goal isn’t to finish more books.

The goal is to rewire how you think.

Here’s how I read now—faster, deeper, and more reflectively—using AI:

STEP 1: Get the book twice.

→ One physical.

→ One digital (PDF if possible).

Read the physical copy first.

Feel the weight of the book. Make reading tangible.

Underline sentences. Jot notes in the margins. Slow yourself down.

Meanwhile, upload the PDF version into ChatGPT.

But don’t use it to summarize.

Use it to argue with the book.

Example:

If I’m reading a clinical research methods book and come across a section on “propensity score matching,” I’ll pause and ask ChatGPT:

→ “Can you walk me through when propensity scores fail to eliminate confounding?”

→ “How would this assumption break down in a real-world PsA study?”

Now, the book isn’t talking at me.

It’s talking with me.

ChatGPT becomes my reading partner—one who questions, pushes, and clarifies.

STEP 2: Read for curiosity, not completion.

Most of us were trained to read like we were cramming for an exam.

Extract key facts. Memorize bullet points. Regurgitate.

But academic growth doesn’t work that way.

The real goal is transformation, not just information.

Example:

When reading a paper on machine learning prediction models in medicine, don’t rush through the methods.

Pause.

Ask:

→ “Why did they pick random forest over logistic regression?”

→ “Would that choice make sense if the outcome was rare?”

→ “What would I do differently if I had the same dataset?”

That’s real curiosity.

Not passively accepting. But actively questioning.

Slow down at the points that light you up.

Let your questions lead the way—not your need to get to the final page.

STEP 3: Digest, don’t just consume.

Most people stop at consumption.

But real learning has 4 layers:

1. Consumption

Read actively. Highlight what surprises you. Scribble questions in the margins. Engage.

2. Digestion

Ask ChatGPT to break it down. Clarify methods. Debate implications.

Example:

“Summarize this clinical trial’s randomization method — and what potential biases could still creep in.”

Force a second brain to challenge your first impressions.

3. Synthesis

Summarize key ideas—in your own words.

Example:

After reading about a new therapeutic strategy, write a two-sentence summary without looking at the text:

→ “This study showed that early TNFi escalation in high-CRP PsA patients led to a 30% higher remission rate without increased serious infections.”

If you can’t say it simply, you don’t own it yet.

4. Application

Apply one insight immediately.

Example:

After reading about how matching on time-varying covariates introduces bias, you might rethink how you’re analyzing your longitudinal EHR data.

Even a small tweak to your own study design is a win.

STEP 4: Write to retain.

You don’t remember what you consume.

You remember what you reprocess.

Example:

After reading a pivotal paper on obesity and psoriatic arthritis, don’t just nod and move on.

Instead:

→ Use ChatGPT to help you reflect. Ask: “What would be a logical next research question based on this study’s limitations?”

→ Use the Feynman technique. Pretend you’re explaining it to a first-year rheumatology fellow.

→ Bonus: Post a 3-bullet LinkedIn takeaway. Share what surprised you.

Writing forces your brain to wrestle with the information.

That wrestling match = deeper retention.

If you don’t create after you consume, you’re just hoarding information you’ll forget by next week.

AI shouldn’t replace reading.

It should deepen it.

More dialogue.

More exploration.

More ownership.

Because every great clinical researcher started as a curious reader—

Not one who raced to the finish line.

But one who stayed long enough with the material to change.

So next time you pick up a paper or textbook—

Don’t ask:

→ “How fast can I finish this?”

Ask:

→ “How do I want this to shape how I think?”

AI can help with that.

But only if you read like it matters.

How are you using AI to learn faster?