Quick Win: The dangerous AI shortcut that quietly erodes your judgment

AI can feel like the ultimate shortcut – no blank pages, no stuck moments. But what happens when you hand over too much of the thinking? 

In this Quick Win, I chat with business professor and author Scott Anthony about how he uses AI as a sparring partner for ideas, and why keeping ownership of the hard thinking is essential if we want to stay sharp.  

Scott and I discuss: 

• Why AI can easily replace the discomfort that leads to original thinking 

• How Scott uses multiple AI models (ChatGPT and Claude) to brainstorm from different perspectives 

• The importance of setting context and personas when prompting 

• Where Scott draws the line between useful AI help and losing his “thinking muscle” 

• A simple test to check if you’ve outsourced too much of your thought process 

KEY QUOTES 

“Use AI to brainstorm, to challenge, to widen the lens – but keep ownership of the hard thinking.” 

“Once you stop struggling with the blank page, you risk losing the very skill that makes your ideas valuable.” 

Check out Scott’s new book Epic Disruptions. 

Listen to the full conversation with Scott here. 

 

My latest book The Health Habit is out now. You can order a copy here: https://www.amantha.com/the-health-habit/  

Connect with me on the socials:  

LinkedIn: https://www.linkedin.com/in/amanthaimber  

Instagram: https://www.instagram.com/amanthai  

If you are looking for more tips to improve the way you work and live, I write a weekly newsletter where I share practical and simple to apply tips to improve your life. You can sign up for that at https://amantha-imber.ck.page/subscribe  

Visit https://www.amantha.com/podcast for full show notes from all episodes.  

Get in touch at [e-mail]  

Credits:  

Host: Amantha Imber  

Sound Engineer: The Podcast Butler 

See omnystudio.com/listener for privacy information.

Episode Transcript

Speaker 1 (00:01):
AI can feel like the ultimate shortcut, no more blank pages,
no more stuck moments. But what happens when we hand
over too much of the thinking. That's what I explored
with Scott Anthony, business professor and author of Epic Disruptions.
Scott has been an early adopter of jen ai, using

(00:22):
it as a sparring partner to test ideas, but is
also clear about the danger. When he once asked AI
to design a university course for him, the output wasn't
just bad, it left his brain out of the process,
and that is where he draws the line. By the
end of this quick win, you'll hear how to use

(00:42):
AI to expand your perspective without outsourcing the critical work
that only you can do. Welcome to How I Work,
a show about habits, ritual's, and strategies for optimizing your day.

(01:02):
I'm your host, doctor Amantha Imba. I know you're obviously
an early adopter of jenai, and I imagine your approach
to prompting and getting the best out of it has evolved.
Tell me how you use it as a brainstorming partner.

Speaker 2 (01:20):
A lot of it is context engineering, you know, making
sure that you put in place the right parameters so
that you are getting a legitimately different perspective on things.
So whether that's taking on a different role, or it's saying,
let's be a skeptic or let's be a supporter or whatever,
trying to make sure that you a set the context
where you're going to get something that is what you're

(01:42):
looking for, which might be something that's as novel as
possible or something that's more fine tune as possible, whatever
it is. The second is multiple tools. My favorite at
the moment are Claude and chat Shept. So I'll play
them off each other. I'lle be simultaneously brainstorming a topic.
I'd be like, oh, the the other one said this,
what do you think about that? And at some point

(02:03):
they converge because all the foundational stuff will get you
to the same place, but they actually will start sometimes
in very different places. And of course we know if
we use the systems, sometimes the exact same prompt will
go in different directions, just depending on whatever is the
magic that's happening underneath the hood. But those two things
are really trying to say, let's push for the personas

(02:23):
and get the context right and then using different models.
Those are the two things that I've found to be
the most helpful.

Speaker 1 (02:28):
I will quite often when I'm working on quite a
big project where I want to get different views, I
will quite often have the same prompt in CHETJPT, Claude
and Gemini and then finish with ask me you know
any questions so you can be ninety five percent confident
and doing a great job. And it's quite interesting the
different questions that I'll get back when it's a task

(02:50):
that could go in several different directions. So I do
love that as a way, and kind of sparring them
against each other I found really useful. One of the
things that I wonder is I think a lot of
people and particularly knowledge workers that you know whose value
is in their thinking use AI a lot. Is I wonder, firstly,

(03:11):
what's going to happen to our ability to think critically
and then what's going to happen to our ability to
maintain good judgment? And I want to know for you
with the thinking critically, how do you think about that
for yourself? And do you worry am I going to
lose the ability to think critically if I kind of
crossed that line where I've suddenly outsourced just a little

(03:33):
bit too much to Ai.

Speaker 2 (03:34):
Yeah, it is something that I worry about a lot.
I worry even more about for my students and even
more for my children. So as you've got then three
generations there, you've got me nineteen seventy five, my kids
who were born between two thousand and five and twenty sixteen,
and then the students, who generally are about thirty years old,
so the average day of birth would be about whatever

(03:55):
that is, nineteen ninety five. It really is being very
clear about the lines. I'm trying to do something that
does represent original thinking. I just can't outsource that. I
can't ask Ai for a first draft of something I tried,
you know when it first came out. Of course, I'm
going to play around with it. So I'm like, okay,
I've got an idea for a new class, please, is
not a curriculum for me? And A the output was

(04:17):
not particularly good, and B my brain wasn't in it.
I hadn't thought about it, I hadn't processed it, I
hadn't struggled, et cetera. So now when I'm working on
a new course, certainly I'll say, Okay, this bit of
it I need to think about some additional readings, or
I want to think about some of the principles I
want to teach here, so I'll get help to sharpen.
But ownership of the integration it has to be by

(04:40):
me or else. That's just a skill that's going to lapse.
So that, to me is the most important thing. Drawing
the line and being really clear about where you draw
that line. There are other tasks that are less critical
where I'm very happy to outsource more so. As an example,
I was facilitating a panel discussion a couple months ago
and that was pretty much Chern and Burn. You know,

(05:00):
here are my panelists. Make me smarter about who they are,
give me a few good questions.

Speaker 1 (05:05):
Task.

Speaker 2 (05:05):
Okay, that's great for material. I can then turn that
into a finished product. But I was totally fine with
first draft being done by AI. But when you really
need to turn on the critical thinking and reasoning, first
draft by AI, I think the research shows is really
really dangerous.

Speaker 1 (05:19):
What I took from Scott is that the line matters.
Use AI to brainstorm, to challenge, to widen the lens,
but keep ownership of the hard thinking, because once you
stop struggling with the blank page, you risk losing the
very skill that makes your ideas valuable. So next time
you're tempted to let Ai draft that strategy, just pause. Instead,

(05:43):
ask it to play the skeptic, or the supporter or
the competitor. Use its responses to sharpen your own. And
if you're keen to hear the rest of my conversation
with Scott, you can find a link to that in
the show notes. If you like today's show, make sure
you follow on your podcast app to be alerted when
new episodes drop. How I Work was recorded on the

(06:05):
traditional land of the Warringery people, part of the Coulan nation.