HomeVideos

We Need To Talk About The AI Cheating Epidemic

Now Playing

We Need To Talk About The AI Cheating Epidemic

Transcript

438 segments

0:00

One of the things that's going on right

0:01

now, I don't know if you all know this,

0:02

but like cheating with AI is rampant in

0:07

universities, in high schools. People

0:09

are just using AI. There all these AI

0:11

detectors. Sometimes an AI detector

0:13

falsely detects something. So like

0:15

teachers and professors are just getting

0:18

demolished because it's something like

0:20

40% of their class uses AI and then they

0:23

go to the administration and they're

0:24

like, "What are we supposed to do?" And

0:26

the administration is like, "I don't

0:27

know. They're an international student

0:29

that's paying us $45,000 a year to be

0:31

here. So, don't expel them because we

0:34

need their money. So, administrations

0:37

are not doing a great job at managing AI

0:40

related things. Professors are on the

0:43

front line of like people are cheating,

0:46

right? And they're like, we're not sure

0:47

what to do. And so I thought this was

0:49

super cool because this professor is

0:51

like, I'm going to go to a cheating

0:52

subreddit and I'm going to ask y'all

0:55

instructor here. Why? Why you guys I'm

0:58

just wondering why you guys do this. It

1:00

seems like y'all are bright bright and

1:01

have clearly spent a lot of time

1:02

thinking through the cheating process.

1:04

Why not just do your

1:05

assignments? I'm not here to say f you.

1:08

My message is this. We went through grad

1:10

school and were professionalized to

1:11

write and do our own work. We take our

1:13

jobs seriously. Many of us see cheating

1:16

not only as an academic dishonesty but a

1:17

personal affront. Why are y'all doing

1:19

this? I I love this response because

1:22

this is sort of what I said. I've been

1:24

an orthopedic surgeon for almost 20

1:26

years and someone was talking about

1:27

this. I thought I would chime in. It's

1:29

pretty wild that people think they are

1:31

beating the system when they are just

1:32

scamming themselves out of the one thing

1:34

they are actually paying for which is

1:36

the ability to handle pressure and

1:37

master a subject. One student mentioned

1:40

that they are too afraid to fail. But

1:41

the reality is that by cheating, you

1:42

have already failed the most important

1:44

test, which is building the discipline

1:45

to show up and do the work when things

1:47

get hard. If you get a degree without

1:49

actually learning the material, you're

1:50

basically walking into the

1:51

workforce/world

1:53

with a fake map and no compass. And I

1:56

was like, damn, that guy is right. You

1:58

guys may have heard me tell this story,

2:00

right? Right. And I'm not surprised that

2:01

this person is an orthopedic surgeon is

2:03

a physician because most physicians

2:06

reach a point in their career where they

2:11

realize it's not about getting the

2:12

grade, it's actually about saving lives.

2:15

Right? So I was on when I was a medical

2:17

student, I was on a boat in Bise and

2:19

there was a guy there who started

2:21

vomiting and I was just a medical

2:22

student but I was like oh crap like this

2:24

guy is vomiting like is he dying? Like

2:26

does he have appendicitis? Does he have

2:28

a ruptured appendix? Did he drink too

2:30

much? Does he have heat stroke? Like

2:31

what's going on? Is this like a medical

2:33

emergency or not a medical emergency?

2:35

Right? So at some point like you have to

2:37

learn the material. And I'm a huge

2:39

advocate of like learning things. So in

2:42

a day and age where AI is like you know

2:45

AI can give you this sort of like

2:47

shallow to medium level expertise in

2:50

anything. But AI does not give you deep

2:52

knowledge in anything. It can't do that.

2:54

I've tried. I've tested. Right? Right?

2:55

If you can I I've asked different kinds

2:57

of AIs a 100 different question not

2:59

literally 100 but like 16 15 different

3:01

questions about different kinds of

3:03

meditation. It gives the same answer

3:04

over and over and over again. So AI is

3:06

kind of like Google version 2.0 where

3:08

Google made a lot of information

3:10

accessible. AI gives a lot of makes more

3:13

information accessible. So the question

3:15

is why do people do this? And this is

3:17

where I'm a psychiatrist. I'm not an

3:19

orthopedic surgeon. So I'm a big fan of

3:21

deep learning. I showed you all a paper

3:23

today about Chinese elderly Chinese

3:26

people who have a victim hero complex.

3:29

Catch I don't know if it's going to know

3:31

that. So why do people do this? We live

3:34

in a world where if you steal a car

3:38

that's worth $30,000, you go to jail.

3:41

You steal $100,000, you embezzle

3:43

$200,000, you also go to jail. You steal

3:46

$100 million, you get a fine.

3:50

You steal a billion dollars, you get a

3:53

bailout.

3:55

This is the world. We live it. Like I

3:58

I'm not even like I want y'all I know

4:00

this is going to be really hard for

4:01

people to do because they're going to

4:02

clip it out of context. They're like but

4:03

but I I want y'all to think about this

4:05

historically. Think about it like an

4:06

alien from outer space. If you observe

4:08

the world that we live in today, and

4:10

there is a good argument to be made for

4:13

actually learning I love it. I

4:14

just made it. But let's not delude

4:17

ourselves into un like forgetting that

4:21

cheating is probably one of the best

4:24

ways to rise to financial prominence and

4:27

success in the world. I know a ton of

4:30

people and if you guys are in academia

4:31

you know what I'm talking about in

4:33

academia there's a bunch of people who

4:36

have engines with post docks with PhD

4:42

students who are grinding with the

4:45

credit flowing to the top right there's

4:47

great labs I bit you know trained at

4:49

Harvard amazing labs there where PI's I

4:52

had an amazing PI one day I'll tell you

4:54

guys the story it's it's something I've

4:55

scripted out about how I became who I

4:58

And I had an amazing PI at at Harvard

5:01

who was just awesome. So there's like

5:02

great people in academia, but there's

5:04

also a lot of frauds in academia.

5:06

There's a lot of people who take credit

5:07

for other people's work. There's a lot

5:09

of people in group projects who don't

5:11

actually do anything and get a grade

5:13

anyway. There's a lot of people who fail

5:15

upward at companies. There's a lot of

5:18

people who show up are charismatic,

5:21

answer questions, but don't do a whole

5:23

lot of work. There's a lot of people who

5:25

when you send them a draft of the deck

5:27

that y'all are putting, they're the ones

5:29

who email it to the partner and make it

5:31

seem as if it is their own work. These

5:33

people get paid. I literally had a

5:35

patient of mine once tell me that their

5:38

boss was like, "You're going to make me

5:40

a lot of money one." Like, "You're going

5:42

to make me a lot of money over the next

5:43

5 years." Literally what the boss told

5:46

him to his face. And you wonder why do

5:50

students cheat? They cheat because it

5:52

works.

5:53

They cheat because it's efficient. But

5:56

what if you get caught? Hey y'all, if

5:58

you're interested in applying some of

5:59

the principles that we share to actually

6:01

create change in your life, check out

6:03

Dr. K's guide to mental health. And if

6:05

you kind of tunnel down, okay, why

6:06

aren't you motivated? And they're like,

6:08

well, there's no point. And if you get

6:10

underneath there's no point, what you

6:13

ultimately find is hopelessness. So what

6:15

the yogis discovered is that what we

6:17

call motivation, they actually called a

6:21

concentrated mind. What's the difference

6:23

between someone who actually does stuff

6:25

and someone who just tries to do stuff?

6:28

So, check out the link in the bio and

6:29

start your journey today. So, let's talk

6:32

about sociopathy for a second. This is

6:35

something that people do not understand.

6:37

When we look at sociopathy, our research

6:40

on sociopathy comes from prisons. Now,

6:44

why does our research on sociopathy come

6:46

from prisons? Because prisons are the

6:48

only place that you have a population of

6:53

sociopaths. Right? If I'm recruiting for

6:56

a study to study something about

6:58

sociopathy, where can I go and find

7:01

2,000 sociopaths? Because in scientific

7:04

research, we want large sample sizes,

7:06

right? We look at populations. We're not

7:08

doing anecdote. Anecdote isn't science.

7:11

I need a large number. So what we do is

7:13

we research sociopathy and we come up

7:16

with a lot of findings on sociopathy.

7:19

But there is a selection bias. Our

7:22

research on sociopathy is on stupid

7:25

sociopaths, not regular sociopaths, not

7:29

intelligent sociopaths. The people who

7:31

get punished for cheating, right? So

7:33

when we study cheating, how do we study

7:36

cheating? How do we know someone is

7:38

cheating? Because they got caught. We're

7:40

not looking at everybody else. So, my

7:44

favorite example of this, this is going

7:45

to be so such a randomass example that

7:47

if you're in the medical community,

7:49

you're gonna you may have heard about

7:50

this, but if you're not, you're going to

7:51

have no idea. So, there's this exam

7:53

called the USMLE step one and step two.

7:56

The exam is rigorous. Okay? So, when you

7:59

graduate from medical school, you take

8:02

these two exams over the course of

8:03

graduation. Your scores on these exams

8:06

largely determine whether you do what

8:08

kind of doctor you become and where you

8:11

do residency. Okay, so this is like

8:13

basically like I want to go to Harvard.

8:15

I want to do dermatology. I want to do

8:16

psychiatry. I want to become an

8:18

orthopedic surgeon. Are you going to

8:19

become a family medicine doc? There's

8:21

competitive specialties, non-competitive

8:23

specialties. Some doctors make a million

8:24

dollars a year. Some doctors make

8:26

$150,000 a year. Some doctors work 80

8:29

hours a week. Some doctors work 32 hours

8:32

a week, 28 hours a week. The other crazy

8:34

thing is that some doctors work 34 hours

8:36

a week and make 800,000. Some doctors

8:39

work 65 and make 600,000. Some doctors

8:42

work 60 and make 150,000. So this exam

8:47

determines how you spend your days, how

8:50

much you put up with, and how

8:52

much money you make. And there was a

8:54

group of people, I think in Nepal, a

8:56

test center started conspiring with

8:58

people to let them cheat on the exam.

9:00

And so these people were getting like a

9:02

a good score is like maybe 230 235

9:05

something like that. Okay, just to give

9:06

you all a sense. And so like some you

9:08

know guy out of Nepal comes out of this

9:10

with like a 270 and there are a lot of

9:13

people who go to medical school in other

9:15

parts of the world who want to come to

9:17

the United States because in the United

9:18

States you make money. Doctors make a

9:20

lot of money here. We don't have

9:22

socialized medicine like the UK or

9:23

Canada where doctors also make a decent

9:25

living there but they don't make bank

9:27

the way that they do here in the US. So

9:29

the brightest doctors like one of my

9:30

roommates when when I was in Boston, you

9:32

know, was a guy from I think Austria and

9:36

his his step scores were ridiculous and

9:39

he was at Harvard and he's like one of

9:41

the like so most of most of the

9:43

prestigious residencies we take mostly

9:45

people who are from the US but we take

9:46

the best and the brightest from the rest

9:48

of the world and so these guys basically

9:50

conspire to cheat on a test. He gets

9:52

like a 270 280 damn near a perfect

9:54

score. So some Nepalese kid gets a

9:57

perfect score on this test. It's a

9:58

one-way ticket to Harvard. And then what

10:01

happens is the test center is like

10:03

great, we we created this structure and

10:05

then many people then suddenly what

10:07

happened is like many people started

10:08

getting 270s, 280s, getting perfect

10:10

scores on this test out of Nepal. And so

10:13

finally at some point someone figured

10:14

out right the USMLE which is the US

10:17

medical licensing exam group figured out

10:21

that someone is cheating because

10:22

suddenly in Nepal like 50% of people or

10:25

20% of people are getting perfect

10:26

scores. That can't be right. So then

10:27

what they did is they voided all their

10:29

scores. Everybody gets a zero. You get

10:31

an automatic fail. And there was a flood

10:34

of posts on the medical school

10:36

subreddits about, "Oh my god, I've been

10:37

failed. I've been failed. I've been

10:38

failed. I didn't do anything wrong. I

10:40

didn't do anything." Everyone's like,

10:41

"Fuck you. You got caught, bitch."

10:43

Right? Cheaters deserve to get caught.

10:45

But what I want you all to think about

10:46

is the first person.

10:49

The first person who figured something

10:51

out gets a ticket to wherever they want

10:53

to go. So why do students cheat? because

10:56

it works. Because it's the way to get

10:59

ahead. Because tax evasion means more

11:02

money. Because it's insider trading.

11:05

Unless you're a member of Congress. Why

11:07

do students cheat? Look at the world we

11:10

live in. Stupid cheaters don't get

11:12

ahead. Smart cheaters are like there's

11:16

so much corruption at the highest. I'm

11:19

not talking about the US and I'm not

11:20

blaming anyone in particular. I mean

11:21

just in general, right? Like so don't

11:22

get triggered by a particular political

11:24

party. Just like look at the world,

11:27

pretend you're an alien and tell me

11:28

there isn't political corruption in like

11:30

50 plus% of countries across the globe.

11:33

I I read this book called Shantaram,

11:35

which is a book about India. And there's

11:37

a great quote in Shantaram. It's a

11:39

fiction book. India is the only country

11:41

with the honest bribe. Bribes are simply

11:43

the way that business used to be done

11:45

there is arguably still done there. I

11:47

don't really know. I don't live there.

11:49

This is why people cheat. And make no

11:52

mistake that I've even seen this, right?

11:53

So in my office I've had people who are

11:55

in banking who are entrepreneurs. It's

11:57

like all these like what's funny is I

11:59

saw a great quote that the Forbes 30

12:01

under 30 has a remarkable pipeline to

12:04

prison. So we have these companies like

12:06

uh what is it? Therronos and FTX and

12:08

like all these like billionaire kind of

12:10

companies. All these like gigantic

12:12

companies where like oh it's actually

12:14

like a Ponzi scheme. Who's that Ponzi

12:16

scheme guy maid off right? And you

12:19

wonder why do people cheat?

12:24

because it's the way to get ahead.

12:26

Now, I want to be clear about something.

12:28

Often times when you explain a

12:30

phenomenon, people think you are

12:32

advocating for it. And here's what I've

12:34

sort of seen. I'm not. So, I am of the

12:36

mind that you should do things honestly.

12:39

I am of the mind that hard work does pay

12:41

off. So, let's be clear about how

12:43

successful cheating is. So if you want a

12:46

reliable way to get into the top 25th

12:50

percentile of success,

12:52

then cheating is not the way to do it.

12:55

If you want to do a good job, you want

12:57

to become a doctor, I don't think you

12:58

should cheat. I think the majority of

13:00

doctors do not cheat. The majority of

13:02

people who are in the top 25%, I don't

13:04

think cheat. I don't think you should

13:05

cheat. I don't think it's good to cheat.

13:07

But just because I don't think it's good

13:09

doesn't mean it isn't effective.

13:11

There's a big difference between what is

13:12

right and is wrong and what works and

13:14

what doesn't work. These are completely

13:16

independent axes that many people have

13:19

trouble separating. So, please God, do

13:22

not think I'm advocating cheating. And

13:24

the other really scary thing that I've

13:26

done because I sort of can't help help

13:29

myself is like when I work with someone

13:31

who's a sociopath, I think about why

13:33

didn't this person get caught? So, I've

13:35

worked in jails and I've worked with

13:37

people who are like in investment

13:38

banking, right? And they have many

13:40

features of sociopathy like they have

13:42

substance use and things like that. And

13:44

there are a couple things that I found

13:46

that are really interesting. So the

13:47

cheaters who don't get caught are the

13:49

ones who aren't greedy.

13:51

This is huge, right? So I want you to

13:54

imagine you're that testing center in

13:55

Nepal. People flew under the radar. You

13:57

can have a couple of blips. It's when

14:00

the signal becomes so large

14:03

that people notice. And like there's a

14:06

ton of corruption out there. There's a

14:07

ton of cheating out there. These people

14:08

are flying under the radar because they

14:10

are below the people who are at the top.

14:13

There's the most egregious corruption in

14:15

the world and then there's like the

14:16

corruption that is underneath that

14:18

doesn't get headlines because it's

14:19

smaller and more boring than the other

14:21

stuff. And I I've had some people in my

14:24

office, man, who are just so good at

14:26

this. Flying under the radar, right?

14:29

Doing things like placing a bribe for a

14:33

toll contract.

14:35

Like it's such a little thing that no

14:36

one thinks about, but now every time

14:39

someone uses this road, this money goes

14:41

in my pocket instead of somebody else's.

14:44

Little bits of corruption, but worth so

14:46

much money. And so, oddly enough, I

14:48

think the the most successful sociopaths

14:51

are the ones who are not overly greedy,

14:54

also keep their narcissism in check

14:56

because sociopathy is correlated with

14:58

narcissism. These are both like cluster

15:01

B things. And the most effective

15:02

sociopath on the planet is one who knows

15:04

when to stop, right? Doesn't keep going,

15:06

doesn't want more, gets their money and

15:08

then is able to hold it. High

15:10

conscientiousness, low neuroticism, and

15:13

leaves their ego at the door. And those

15:16

people you will never see research on.

15:19

You'll see research on the people who

15:21

grew so fast, the Forbes 30 under 30,

15:25

that everyone was looking them and they

15:27

got scrutiny. And so I think if you're a

15:29

professor and you want to know why do

15:30

students cheat, it's because it's one of

15:33

the most effective tactics.

15:36

Think about what cheating is.

15:39

It's getting the fruits of labor without

15:41

the labor. You know what we call that

15:43

when we work in corporate? Efficiency

15:47

optimization.

15:49

Minimizing the denominator, maximizing

15:51

the numerator. And the only time it's a

15:53

problem is when you're stupid enough to

15:55

get caught.

Interactive Summary

The video explores the pervasive issue of cheating, particularly with AI in education, and its broader implications within society. The speaker, a psychiatrist, highlights the struggles of educators facing AI-driven cheating and administrative reluctance to act due to financial motives. He contrasts the 'stupid' cheaters who get caught with 'smart' cheaters who succeed by flying under the radar, often operating within corrupt systems. Using examples like a surgeon's perspective on the self-deception of cheating and a medical exam scandal, the speaker argues that while he does not advocate for cheating, it is often perceived as an effective strategy for personal and financial advancement in a world where large-scale corruption can go unpunished or even rewarded. He identifies key characteristics of successful cheaters as being non-greedy, managing narcissism, and knowing when to stop, allowing them to remain undetected and achieve their goals.

Suggested questions

5 ready-made prompts