Dr. Frank offers a sharp reality check on the limits of AI, noting that algorithms can manage symptoms but fail to dismantle the core psychological lies of addiction. True recovery requires a human touch to challenge internal beliefs that AI often inadvertently validates.
Deep Dive
Prerequisite Knowledge
- No data available.
Where to go next
- No data available.
Deep Dive
Quitting Weed With CHAT GPT Might Be A Bad Idea...Added:
The inspiration for today's video comes from a client of mine that I've been working with. This is a highlevel medical professional who has a indepth knowledge of neurology and they work in the mental health space. And this uh individual is struggling with a cannabis substance use disorder. And on our call, we had a long discussion about triggers and trying to understand triggers when it comes to consuming cannabis and uh cannabis substance use disorders. And this individual had been using an AI platform like Claude or Chat GPT in hopes of gaining a deeper understanding of why they are triggered to consume cannabis, why they've had a history of relapsing, and what they can do moving forward. Now, as a disclaimer, I'm a huge fan of using AI models when it comes to helping people quit substances, but you have to do it correctly.
Otherwise, you could wind up with the the completely wrong information if you're not prompting the AI model correctly. If you're not pulling data from the correct sources, it could send you down the the completely wrong rabbit hole. If you're new here, my name's Dr. Frank. I personally struggled for several years with nicotine, cannabis, binge drinking, adult media content, and stimulants. And we now have a science-backed method that has a 98% client success rate at helping people break the dependency in 18 days or less.
Our method is designed specifically for business owners, entrepreneurs, and other highlevel professionals. People who want to do this efficiently, effectively, privately, and most importantly, without relapse. I put a link in the video description and pinned comment where you can learn more about that. This individual had gone on to uh chat GPT and they were saying, "Hey, I'm trying to figure out why it is I'm triggered to keep consuming cannabis over and over again." And the AI model and him uh figured out three identifiable triggers. Okay, trigger number one, which is one I hear all the time from people, was boredom. And I'll say the AI model did a really good job at helping this individual identify what what is behind the boredom. Is it you don't have enough activities going on?
Is it that your days just aren't busy enough? Is it that you're bored in your career or in your relationships? And essentially what they had identified was that the boredom was rooted in a lack of meaning in day-to-day stuff and a lack of purpose. And Catch GBT had said, "Look, here's an identifiable trigger.
You're experiencing boredom that's associated with a lack of meaning in your day-to-day activities, parenting, career, hobbies, whatever it was." and the AI model had then uh dissipated different ways for this person to find meaning in their current day-to-day activities. And I think that's great.
That's really good. It's cool that it got that in-depth insight behind boredom. The secondary trigger that it identified was discomfort. And the AI model kind of poked and prodded at this person and it got them to uh find out that the discomfort was more so not like a physical pain or a physical discomfort that they smoke cannabis, but it was more so related to social anxiety and social discomfort. And this person says, well, you know, I smoke weed because it makes me more comfortable socially. I binge drink because it makes me more comfortable being social. And the AI model basically was like, great. So what we need to do is we need to work on finding you other methods to become less socially uncomfortable and it made a series of recommendations for him.
Lastly, the AI model determined that a trigger was what I refer to as the kind of f it mentality. F it. I'm just going to go smoke weed because why not have some fun? Life is good. I'm successful.
I have a family. F it. Who cares? Let's have some fun. let's go smoke weed. And he had not gone any further into exploring that trigger with the AI model at the point he was talking to me. So I said to him, I was like, this sounds great. You guys, you know, you and this AI model identified some reasonable triggers that I hear constantly as someone who's worked with numerous people over the last seven years, eight years now doing this. But I said there's one big problem with each of the things that the AI model told you. So let's look at the first example of boredom. So it identifies that his boredom is related to a lack of meaning. And I said to him, you know, it's it's great to to look for deeper meaning in your day-to-day day activities, to identify deeper meaning in raising your children and pursuing your career, but where the AI model fell short was it was inadvertently validating his belief that consuming cannabis provided meaning to these experiences. And this is a little nuanced, but it's insanely important. If you're trying to quit smoking weed and you don't feel that you have a purpose or you don't feel that you have enough meaning, even if you are successful, you have a good meaningful career, you have children, you have a family, you have all that stuff, right? People can still feel like they're lacking meaning in life. And the AI model was basically saying, "Yeah, you're right. You're not finding meaning in things, and that's why you smoke weed." So it was inadvertently validating that somehow cannabis is providing meaning to these mundane or day-to-day activities and that we need to find deeper meaning in order to quit consuming cannabis. And I had said to this person, listen, it's it's actually much much simpler than that. Yes, it's great to find a deeper meaning and find a deeper purpose in what you do. I'm a big fan of that. I have tremendous purpose in my day-to-day life and that really does it fills me up. I feel great about it. But that's not a requirement for quitting cannabis. What we really need to focus on is your belief that consuming cannabis somehow provides meaning. Because if we can devalue cannabis, it makes quitting significantly easier. And now if we can devalue cannabis, not having meaning, experiencing boredom no longer becomes a trigger. And I explained to this person, boredom and a lack of meaning in your day-to-day life is only triggering to consume cannabis if you feel that cannabis somehow resolves that problem, a lack of meaning. And I had said to him, listen, when we first spoke, you made it very clear to me that you don't find meaning in consuming cannabis. In fact, you made it very clear to me that cannabis takes away from the things in life that you actually value, that you find true purpose in, that you find true meaning behind. So, I posed the question, if that's true and you are bored and it's related to not having purposeful, meaningful things that you're identifying day-to-day. If that's true, how is cannabis going to resolve that for you? Because you would only be triggered to consume cannabis in that scenario if you felt that cannabis could provide you meaning. And he was like, "Oh, okay. That makes sense." So what we're doing here is we're taking what the AI model said and we're building on it appropriately. And this is a big thing to build on because now the conversation isn't so much about you got to do all these extrrenuous things to find meaning. It's more just, hey, listen. Yeah, do all these extrrenuous things, but you don't need to do that to quit. You just need to acknowledge that cannabis isn't actually providing you with meaning in the first place. It's not solving this problem for you. And that's why you go back. That's why you're triggered to use it in this instance. Now secondarily we talked about discomfort and it was identified by the AI model as a social discomfort that he was experiencing. You know kind of awkwardness in social settings, social situations and the belief that weed or alcohol helps with that. And I had said to the person okay so you know AI said look we got to work on social things. Maybe take a class on becoming more sociable whatever all these things which I think is awesome. I actually had to do that at one point in my own life.
I had a big problem making direct eye contact with people and cannabis and alcohol kind of helped me with that quote unquote. And I actually had to go out of my way to take like classes and do workshops on making direct eye contact. Okay. So I I could really relate to this with this individual and I said, "Well, here's the problem here.
The AI model is validating your belief that cannabis and alcohol are somehow helping you alleviate the social discomfort. [snorts] So even though the chat GPT didn't understand what it was doing, it was actually just reinforcing this person's belief system that alcohol or cannabis was a solution to their social discomfort, which in fact it was far from a solution to that. So I had pointed out to this person, we had kind of workshopped through, hey, I want you to identify all the ways that cannabis is actually causing discomfort in your life, not alleviating it. And I want you to identify all the ways in which cannabis is causing discomfort that's actually piling onto that social discomfort, not solving it for you.
Because now what we're doing is we're devaluing that perceived benefit of that drug and we're removing the necessity of it. See, the AI model was good. It gave good insight on how to fix this problem, but it didn't do anything to address the fact that cannabis was not ultimately the solution to this problem. In fact, the AI model was reinforcing the belief that it was the solution to this problem and it was kind of encouraging this individual to find different habits, different routines, different ways to work on this. And now what the AI model is doing is inadvertently making quitting a substance about habit replacement, routine replacement, social practicing, whatever. And [snorts] quitting drugs has nothing to do with that. It has to do more with addressing the core belief that somehow cannabis is helping me socially, somehow alcohol is helping me with socially. And if we can destroy that core belief, then those social situations are no longer going to be triggering quote unquote to consume the substance. Now, for problem number three, AI and him had identified you have this effort.
I'm just going to go smoke weed to have some fun. And I can tell you had he pursued it further, the AI model model would have said, "Well, what do you do for [clears throat] fun? What do you find that's entertaining? Go do more of that stuff." Again, making quitting drugs about habit replacement and behavior replacement, which it's really not. So, I said to him, "Well, why are you talking to me right now if consuming cannabis is fun? If it's fun, we wouldn't be having this conversation. If something is truly enjoyable, you don't want to stop doing that thing." thing.
And I gave him the example. I said, "I have a lot of fun going hiking outside on a beautiful day." I don't get home and say, "Oh my gosh, I never want to go hiking again. This is terrible that I'm doing this." No, I'm excited. I can't wait to go on another hike. I'm excited to do more of that. So, I said to him, "Well, okay, there's two parts here. You got this effort mentality. So, one, can't you just flip that mentality around? You say, "F it. I might as well smoke. Who cares?" Can't you just as easily say, "F it. I might as well not smoke. What's the difference anyways?"
That was that was advice number one.
Stupid simple. Just a quick quick flip of framing. Advice number two that I gave him was, and this is really the core of this trigger. You feel, you have the belief that consuming cannabis is fun. Yet here you are speaking to me, working with me, trying desperately to quit consuming cannabis because you don't want to be doing this anymore. So identify for me where the fun is in this behavior. Is it actual fun or are you just under the illusion, the delusion of fun, the perception of fun? Because that's what you've identified smoking weed with for so many years. You get high, you associate that high with fun.
Is it actually giving you that though?
Right? And this is kind of those light bulb moments where it's like, wait a minute, you're right. Why am I talking to you if it's fun? because that that thought process would only be triggering if you believed it to be fun. And I said to this individual, listen, I am all for using AI models to help you quit substances. But in this example, AI missed the most important piece to all of this, and that's addressing your own internal beliefs about what you think that drug is giving you, your own internal values about what your relationship is with this drug. And it's not just AI that incorrectly did this.
The reason why AI isn't as good at at this as I am is because it's pulling data from whatever's out there on addiction recovery. And that data is largely rooted in 12step programs, articles published by rehabs, online therapy writeups, and things like that.
And this is very problematic because one, a lot of that stuff is not evidence-based. Your 12step program is not evidence-based. Okay? A lot of these things have insanely high failure rates.
Rehabs, what 10 to 12% success. Therapy, I think 15% success. 12 steps, I think, like an 8% success rate long term. So, it's pulling data from sources that are not even good to start. And that's the feedback that it's giving the person.
And that's why the AI platform was overlooking the root cause of this person's substance use disorder. And in order to see this, you got to be entrenched in this. I've been I suffered from addiction for nine plus years. I've been coaching now for eight years, right? Like I'm deep in the trenches with this stuff. Thousands of conversations with people, working with people directly, one-on-one, hundreds of individuals. And again, if this is clicking for you, we have a scienceback method that has a 98% client success rate at helping business owners, entrepreneurs, and other highle professionals break the substance use dependency in 18 days or less. We can do this efficiently, effectively, and privately. Maybe you're a public figure.
Maybe you're a physician yourself. Maybe you're an attorney. Maybe you're a micro or macro influencer and you need help with this. I put a link in the pinned comment in the video description. We can do this privately. No one has to know.
Okay? The only trigger is you. It [snorts] has nothing to do with these external factors that unfortunately AI made it all about. AI made it all about.
There's all these external factors and you have to do X Y and Z to try try and control them. And the solution is actually much easier than that. You just have to control the thoughts in between your two own ears. You just have to identify that you are the trigger, your belief system, your values, your relationship with the drug. That is the only trigger. Because if you put triggers onto external things that you cannot control, you will constantly be running from substance use. You will constantly be making this about behavioral changes and habit replacement. And you will eventually burn yourself out, get fatigued, and there's a very high chance that you will return to problematic substance use. But if you can identify that fixing this is actually much easier than that. It's addressing your own internal belief systems, your own internal behaviors, your own internal values. Oh my gosh, quitting becomes a thousand times easier and you can finally stop running from triggers. I'm going to do a follow-up video to this where I'm actually going to have a live conversation with Chat GPT or Claude. Maybe I'll do it with both of them. and I'll show you just how flawed these AI models are when it comes to giving appropriate feedback to stopping the use of a substance. Now, some other shortcomings of AI. It it obviously can't do a medical detox for you. We have physicians that we work with and they'll work with people all around the world where we can actually provide inhome medical detoxes if that's something that you need, a supervised medical detox. And the reality is AI doesn't know what it needs to be asking you to get the information. Because when it comes to quitting substances, it's more about like when it's me helping someone, it's not what I'm telling them that's important. It's the questions that I'm asking them cuz that's what's going to give me the information to help them. Whereas AI is predominantly you prompting AI to tell you something.
Whereas it really needs to be much more about extracting information from you, not you extracting information from AI.
Because without AI having your full history, without AI having clear prompts on how it should be advising you, you can wind up down a really, really bad rabbit hole with it. So that's my thing.
But when you have the correct prompts, when you know how to provide it with the proper information, AI can be insanely effective. And this is something that that I teach people inside our one-on-one uh coaching programs with us.
So, if that appeals to you, again, check out the link in the pinned comment or the video description or just follow me right into the next video where I talk a little bit about relapse and why relapse is not always a bad thing. Cuz it's not
Related Videos
OpenHuman VS Hermes AI: Who Wins?
JulianGoldieSEO
285 viewsโข2026-05-29
Long-Running Agents โ Build an Agent That Never Forgets with Google ADK
suryakunju
142 viewsโข2026-05-30
This computer is made from real human brain cells. And you can buy it.
Talktmsmedia
3K viewsโข2026-05-28
BREAKING: Microsoftโs New Image Generating Model Beat Out GPT 1.5 and Nano Banana 2
aimmediahouse
122 viewsโข2026-06-03
I Made the Same Anime Fight Scene in Every AI Video Generator
NobleGooseAnime
295 viewsโข2026-05-30
Nvidia Bets Big On AI PCs | New Chip To Power Windows Laptops | Technology | AI Updates | N18S
cnnnews18
3K viewsโข2026-06-01
I Tested NEW Opus 4.8 on Four Projects (Updated LLM Leaderboard)
AICodingDaily
298 viewsโข2026-05-29
3D Platformer Update - NO CAPES
SolarLune
294 viewsโข2026-05-30











