My Podcast

In a hurry? Don’t feel like reading?

Listen to my podcast by clicking here.

Table of Contents

1. Preface

2. Main Article

3. Readers’ Favorite

4. Stuff

Preface

This week, we've started to witness the first signs of spring.

Don’t get me wrong, it’s still freezing when I commute to work, hovering between minus 10 and zero.

But we now get sunny days here and there, reaching up to 15 degrees.

The spring also signifies approaching the end of my fellowship, with 80% of it already behind me.

The focus now shifts towards mastering autonomy—handling surgeries on my own, managing the OR, overseeing inpatient care, and sharpening my decision-making skills.

After two decades, my formal training is finally nearing its end. Now, a new chapter is unveiled. That has been quite a ride. Time to come home.

Last week, I wrote on Readwise Reader, an app that has revolutionized my reading. You can read about it here in case you missed it.

It has some cool AI capabilities that will summarize documents and books for you, and even YouTube videos!

But it’s not a dedicated AI tool. It’s a reading tool. It enhances the reading experience without taking over the intellectual portion. That’s why I like it so much.

For more demanding AI tasks, there's a clear leader: ChatGPT.

However, this brings us to a crucial question: As doctors and scientists, should we use AI?

Let's dive in.

Seven-seven, here we go!

Main Article

It’s been a while since ChatGPT has stormed into our lives, but whether (and how) doctors and scientists should use it deserves a separate discussion.

Are we just using it to complete repetitive and technical tasks, or are we risking our professional integrity and the essence of our expertise?

To answer that, there’s a simple exercise I want you to do right now.

Imagine ChatGPT as a real person sitting next to you.

Take a good look. Who is this person?

First, this person is your employee who earns 20$ a month and works 24 hours a day for you. Not a bad start.

Second, this employee is efficient, quick, and will never say no.

But is this person smarter than you?

No.

This person just knows how to search the internet at blazing-fast speed.

Is this person original?

Not a chance.

By definition, ChatGPT will never be revolutionary. It will present you with the most common, average solution.

And in science, that’s bad.

Is this person reliable?

MOSTLY, yes.

Sometimes, NO.

And in science and medicine, that’s bad. VERY VERY bad.

“So what are you saying here? As a doctor or a scientist, I shouldn’t use ChatGPT?”

Quite the opposite! I think it’s a great tool. To a point.

As you wouldn't let the person in the room next to you replace the depth of your insight, creativity, and seasoned judgment, the same goes with ChatGPT.

So here are my 6 tips for using ChatGPT:

🔵 1. Preserve Originality

Don’t ask ChatGPT to write something from scratch.

In other words, if you wrote one sentence and ChatGPT responded with a long text, you just gave away your intellect and creativity.

🔵 2. Edit, Don't Rewrite

ChatGPT is a great tool for editing, helping to refine structure and grammar without altering your distinctive voice.

If you end up with a text different from your own, that’s not editing. That’s ChatGPT taking over.

🔵 3. Recognize the Style

ChatGPT has a recognizable style. You may think nobody will notice, but they will.

That’s especially true for doctors and scientists. If you’re hiding the fact that you used ChatGPT to write for you, it’s a sign you went too far.

🔵 4. Innovation Over Automation

True innovation in research and medicine cannot be automated. ChatGPT will ALWAYS output the average response. AVERAGE.

Being innovative and creative rests solely with us.

🔵 5. Moderation is Key

Think of ChatGPT as an advanced spell checker. When you rely on it, it will diminish your professional voice and impact.

Very bad for people of science.

🔵 6. Social Media

Unlike medicine and science, the originality of your posts on social media is not that important. Using ChatGPT to write the posts for you is risk-free, and probably nobody will hold it against you.

But it’s boring. So boring.

It seems that nowadays, ChatGPT generates every other post on social media. The result? All the posts look the same — long, fluffy, and robotic.

When you write your own posts, people will notice and interact more with your content.

Your personal touch is (still) irreplaceable.

Readers’ Favorite

Afraid to Speak in Public?

In issue #54, I name 3 reasons you should do it anyway.

From the value of public speaking through networking to gaining respect.

Missed it? Read it here.

Stuff

Ridge wallet

That wallet has been in my pocket for 5 years.

It still impresses me, and I don’t see any reason to replace it anytime soon.

Top it off with a lifetime warranty and excellent customer service. Check it out at Ridge.

Epilogue

If you received this newsletter from a friend and would like to join Friday's Digest, visit https://newsletter.shaysharon.com

That’s it for this issue.

Hope for better times.

Shay