Transcripts
1. Class Trailer: Keeping up with an
understanding AI is one of the most important skills in
modern day creative work. And I believe firmly,
as a creative that we need to understand
what is happening in the AI space so that we can continue creating at
the highest level. Artificial intelligent. Is
artificial intelligent. This will be the greatest
technology humanity has yet to be. AI as the most
profound technology, more profound than
fire or electricity. Now, in this course, we're
going to learn about how to understand AI and its connection to large language models, which you might
have heard through chat GBT, cloud or perplexity. A lot of these big
language models is essentially all categorized
into the AI group. Now, in this course,
I'm going to show you and break down
these concepts of how a large language
model works so that when you're prompting and
when you're working with AI, you'll understand what is
happening behind the scenes. In this way, you'll gain a foundational
understanding of what AI is and how you can apply it to practical applications.
Hi, I'm Arnold. I've been a creative
director and in the advertising space for
over the last decade. And in the last year alone, I've spent all my time devoted to working with
AI and understanding how AI can be worked and
improved for a creative. And over the last year, I've been focused deep on how we can leverage AI as a creative, because I believe if we're
able to understand AI, we can bridge the gap from
where our creative vision is, where we currently are. Some of the course takeaways that we're going to cover are, understanding what AI and LLMs
are and how they function, Insights on the impact of AI
on the creative industries, practical skills and
utilizing AI tools, how to integrate AI
into your workflow. And finally, we'll finish
off with confidence in your ability to discuss and
understand AI concepts. This class is tailored
towards anybody that needs to use
creativity in their work. If you're an entrepreneur, if you're a designer, if
you're a videographer, if you're a filmmaker, if you're an artist, AI
can help you in many ways. And so this class is
to help you understand how you can leverage
your knowledge and augment it using AI. So if you're as excited
as I am to embark on this journey into the world
of AI, then let's begin.
2. Introduction to Class: Welcome to the class. So let me introduce you to
what we are going to talk about in this class
and what I want you to take away after you
finish this class. Now, the goal is for AI to be demystified by the
time you finish this class. And what I mean by that
is that you'll have a solid understanding
and approach to AI. And to understand that it's not artificial intelligence
but more of an information retrieval system that augments your abilities. In other words, AI or what
the current rendition of AI is is not going to replace jobs because it doesn't
have the capabilities to. But it is very good at retrieving information
that was taught to it. And your job as the
creator is to come up with the prompts and to
guide this system to retrieve that information so that it can output something for you to help you get the
solution that you want. So in the class,
we're going to go through the fundamentals of AI, so you understand the
mechanics behind it. I'm not going to get
super technical, but I'm going to describe an overall technical
understanding of it so that you can get an
idea of what's happening. And that way, you can have a better understanding of its applications
for your purpose. We'll also talk about
the limitations and the possibilities as we go
into a more hands on section. So you can watch me work on AI and how I do my reiterations as I'm continuously building on an AI platform to get
an output that I want. And then finally, I'd
like to see what you finish with and what you
present in your project. So let's move on
to the next video. I'll introduce you
to the project, and then where are you
going to get started.
3. Project: So, the project for this
class is for you to use AI and to augment a certain topic that
you want to talk about. So this is very important
for content creators. If you have a topic
that is showing some traction and your
audience seems to like it, why not dig deeper and continue making videos
on this certain subject. This includes making
video content or blog content or even just writing out
promotional material for your brand or
business or company. So for the purpose
of this class, I'm going to go through
a hands on practice where I had this video that got hundreds of
thousands of views, and I know that the audience wants to
know more about this. So, how are we going to use AI to dig deeper
into this subject. And then from there,
I can generate ideas or I can generate different topics that I wanted to cover, or I could get a better
understanding of this topic so that I could
communicate it to my audience. So using AI, we're
going to augment and create more with this subject
that you're working on. So what I want to see from you, you're going to use AI and you're going to
converse with the AI, and then you're going to
come up with a couple of different ideas that you hadn't have thought
about before. And I'd like you to share
it in the project section below and write down,
This is my subject. And these are some things
that I hadn't considered until I started using a
certain language model. Now, to make it easy, you
can use free language models like Chat GBT, cloud, or Lama. There are paid versions to this too that are more powerful, but they also have
some free versions that anyone can use. All right. Up next, we're going to talk
about the fundamentals, technicalities behind what is happening as you're using
these large language bottles. So let's move on to
the next lecture.
4. Understanding Large Language Models (LLM's): In this section, we're going to learn about the fundamentals of how the technical parts of a
language model works in AI. And to start with that, we need to understand
something called X, which is retrieval,
augmented generation. Basically, most language models these days use something
that is like this, and this system is how your
information is processed, like the prompt that you give
it is processed and goes through this whole cycle and then spits out
an answer for you. Of course, this does get
a lot more complicated, but I'm giving you an overview so that you can
understand what is happening as you giving
this AI your text prompt. And as it breaks down
this text prompt, it sends it off
to the next step, then from that step it
goes to another step, and then finally, it
goes back to you. So to start off, we send off these props to the language model
that we're working with. And these are all
tokens in a sentence. And as this system
retrieves these tokens, it starts running information and reading this
token and seeing, Oh, maybe this word has more weight in this sentence
versus another word. So, for example, if your
prompt is something like summarize this transcript. There's three words in there, and the two important
words summarize and transcript are going to be
weighted higher than this. So as the sentence gets
put into the machine, the machine is
going to know that there's these two
important words. So they all have
their own algorithm, and they all have
their own, like, different weights of which terms mean more and which
terms end up doing what? So every LLM is different. That's why I always
recommend that if you're trying to use AI to work with many different AI
models and then to find which one is
perfect for your needs. Because, for example, if you're writing maybe Cloud
is a better fit, or maybe if you're trying
to come up with topics, then Chat GBT might
be a better fit. So it really depends on
which is the best fit for you for the specific purpose
that you are using it for. And so that's the
first step retrieval. Once you send all
your information through to the machine, this AI machine, it's going
to retrieve information. It's going to go up to the data bank that it has or whatever learning
data that it has, and then it's going to
retrieve information. Now, that leads us to the next
step, which is augmented. And that's where it
gets all this knowledge that it has and integrates it together with what we
are asking it to give us. So like I mentioned, it
has this knowledge base, which is based on how
each language model determines certain words or
which one is more important, which one is less important, and how it can predict a
good output feedback to you. So it goes through
this knowledge part, and this is where it augments what you are asking it to do because it's retrieving a lot of information and processing
all of this information. And then finally, that leads to the third step,
which is generation. And this is the answer
that it gives us. And the generation part is
taken everything from before, the information
that you gave it, it weighs out which
part is important, which part is not important? It digs into the data
that it was fed, which could be data that
was specifically taught or data that is fine tuned for a specific
model that you had, like, trained to understand. And so it takes all of the
things from the steps before, and then it generates
out an answer for you. And then from there, it
goes back to the user, which is you because
you got the answer now. And then once you
get this answer, you need to confirm if this is the type of
output that you want. And if it's not, you
got to go back and then you have to reiterate and test out different prompts
so that you can guide the ag system so that it can retrieve and augment and generate out this
answer that you want. And so in a nutshell, that's how AI systems
work these days. Of course, we can
dive in a lot more and talk about how
neuro networks work and how that implements
into AI and how AI has certain biases
and ethics issues. This is infinitely debatable, and we can go really deep. But for the purpose
of this course, and for you as a creator, this should give you
an understanding of what is happening as you put in an input and as you retrieve an output
from the system. All right. Moving on.
The next lecture, we're going to talk
about a formula for you to consider
as you're writing out prompts for certain
AI language models. Okay.
5. Choosing the LLM for YOU!: In this section,
we're going to talk about choosing the LLM for you. There's a misconception that whatever LLM language
model that you're using, will get you to the
goal that you need. And essentially,
you could do that, but there are preferential
language models per the use case
that you're using. So, for example,
it's basically like comparing maybe an Android
phone to an iPhone. And the difference is that even though some language
models might outperform, other language models
in their benchmark, there's different
individual aspects of that language model where it returns an output to you that is preferable to
what you want to make. So in this section, we're
going to go through Chat GBT, and we're going to use something else and compare it side by side so that you can see
what the returns are, the outputs that we
are getting from it so that you can choose which
one is the best fit for you. Now, I'm going to
explain why I like or prefer a certain response
over another response. And if I were to use it in a certain use case of
content production, why I would choose this
one over the other one. So let's jump in
and get started. Alright, so now we are on Versa. Verso is a good platform for us to compare LLM models side by side so that we can see exactly what our input
generates and gives us. So here we're going
to use Cat GBT 3.5 and also Google Gemini. I'm going to go through and talk about this prompt that
I'm going to enter, which was me doing
research a few days ago. So as I was doing
research for something, both different LLM models
gave me a certain response. So I'm going to
discuss why I like certain responses
versus another one. So let's get started. Now, first, I was exploring Bowman and Deals for frame
models of leadership. I use LLM models a lot to dig deeper and learn more
about a certain subject. So we'll start with this. What is Bowman and Deal's for
frame model of leadership. So they both came out with these generations. So
let's look through them. Now, we notice right away that Open AIs Cat GBT
is more compact, whereas Gemini gives us a bit more information
to go off of. So if I were to study the
subject, for example, I would look and see and see
what there is to consider. Now, if I want to just brush over this and to
just see what's going on, like, this is a bit
much to read, right? Like, if I'm in a conversation
with someone and I just want to get the gist
of what this is about, I might ask something like, Hey, can you please
simplify this? So I would say something
like please make it simpler her model. Like like a sentence each, Now, I have to say I do like Geminize breakdown here
where it talks about focus, key elements, and
leadership styles. But right now in, like,
most of the time, I'm mostly just researching and just trying to get
an idea of what is happening or what even is the subject so that
I could start thinking and connecting more about what I know we're
hunting creating, right? So I want to know
about this subject before I can consider
more about it. So I want to make this simpler. And as we asked it
to make it simpler, we have two approaches
here or two outputs here. And so, as I'm
reading through this, I see some that I
like, for example. Maybe I like the symbolic
frame of approach as I'm talking about leadership or as I'm trying
to lead a group, a company or a club. So I'll ask it to tell me more about the
symbolic frame model. And so from here we can see
on Chat GBT, the left side. It's straying away
from being simple, like we asked it earlier, and on Gemini, it's still giving us a very simple,
straightforward answer. Now, depending on the question, this is just asking
for it to expand more. So it's understandable that this gives us a long response. This way, I'm learning
more about the subject, but what if I wanted
to apply this to something that is more
tangible to my situation? Now, I want to ask
it to frame this in a way that is more towards a startup founder
or business owner. And from here, we start to
see different parts and more specific parts of
how this framework works. So on Chat GBT side, it talks about how we can
study organizational culture, conduct cultural assessments,
define core values, create meaningful rituals,
tell compelling stories, encourage symbols of identity. All of these are things that potentially a business owner, start up founder could implement into their
management style. Now, ago Gemini, it breaks
it down into a couple of different ways
where it starts off by telling us how we can learn And then how we could apply it. And then finally, what are some benefits of
taking this approach? So from just these two, we already see that
with open AIs chat GBT, we're getting more
of a response that is straightforward towards
what we asked for. And on Gemini, I'm
getting more of a overall view that can give me more information about
what I am learning about. Now, if I were to just
do an article or if I were to just make a video
about management styles, maybe the Chat GBT response is better because here
it's more straightforward, and I can just talk about these points that are listed
on the Chat GBT side. But if I were to
learn and incorporate this into my own start
up or my own business, I might consider Gemini side
so that I can think and see how this affects different
aspects of my business. Now, I'm going to follow
up with two questions. I'm going to ask it to give me some successful leaders who have used symbolic frame
effectively and try to identify the
specific strategies and techniques that they use to communicate their
visions and values. And then I'm going
to ask what are some opportunities that
I can use to create symbols and rituals that reflect my company's
cultures and values, and as they listed, such as team
building activities, award ceremonies,
or company models. So I'm going to ask it to give me some suggestions for
these two questions, and we're going to see
what it comes up with. Alright. So the first part, we have the successful leaders, and they both, for some reason, mentioned Howard
Schutz and Steve Jobs. So these both give me an understanding of the
scene of what's going on, shows what these people have done in their
respective companies. And finally, at the
end, third suggestion, they added Tony Sha
for examples and Indra ui for Pepsi Cs
for the Pepsi company. Now, I was more interested
in the opportunities that would implement that
style of management. And then that was the second question that I had prompted. And then here it
gave some ideas. So on left side with Chat GBT, it talked about team
building activities that we could possibly
do or award ceremonies, or different company
models and slogans, office design elements,
social events. And then on the
right with Gemini, I talked also about team building activities,
award ceremonies, slogans as well and then
also goes into traditions, symbols that they can use or how internal communications
could be done. So now that you see that
both platforms gave you a variable type of response based on even the
same prop that we gave it. And on one side, we saw
that with Chat CBT, I was able to get more of a direct answer that could
help me just brainstorm ideas, whereas on Gemini, I was getting answers that gave me an overall look of the
whole situation. So depending on your situation, depending on what
you're trying to build, as you're doing content, it might be helpful
to be using Chat BT. But if you're trying
to brainstorm your company and brainstorm
the direction of how to lead, Maybe Jim and I might be a better choice in
this situation. All right. No up next,
we're going to put this more into practice. Okay.
6. How to make QUALITY content: So with the prevalence of AI
and the easy access to AI, there's going to be an influx
of low quality content. And what I mean by low
quality content is if somebody goes on
Chat GPT and says, Hey, write me an
article about surfing, takes that entire article and
publishes that as a blog. Of course, there's going to be so many missed nuances because they have
solely relied on AI. So in this section, we're going to talk about how to create good content using AI because AI is not here
to do your job for you. It's here to augment
your intelligence. At least that's what I think AI stands for augmented
intelligence. So we'll first get
that out of the way. The misconception that
artificial intelligence is going to be here to replace us and to do everything for us. No, I don't believe so. I believe that it
will have a lot of information that
is accessible to a great amount of people,
which is amazing. But with that comes a lot of people taking
advantage of that, having really poor quality
stuff being posted online. And so in this section, we're really going
to dive down into how to make good
quality content. And so how I like to start with this is to start with
something that you know, or If you don't know about something,
you just heard about it, start with that and then build into something that you know, because as you're providing content or as you're
making content, this should be a benefit
to whoever is viewing it. So make sure you
do your research, make sure you're qualified, make sure you know what
you're talking about. And AI is very good at that
because it can help you do your research and can help
you verify your research. And then it could also
give you ideas on what different angles and different considerations
you have. So the first step is to start with
something that you know, which subject that
you're talking about. And if you're a content creator, you probably already know
which niche st you're in, what value you provide. So you start with that path. So, for example, this video
that I was doing that got a couple of hundred thousands of views and a lot of comments. This was based off of
my channel where I was talking about mental
health. And wellness. So when I was making this video, I had that framed into my mind. So once you have
this topic framed into your mind, in this example, this topic is about blue
zones where there are people that are living very
healthy living long lives, and they are, for the
most part, happy. So I was breaking down
the different parts about these people's lifestyles that contributed to their well being. The longest living
people in the world live in places considered blue zones. And these blue zones are
Okinawa, Japan, Sardinia, Italy, the Coa, Costa Rica, Ikaria, Greece, and Low
Melinda, California. These people are known
to live well into their 90s and are happy
throughout most of their lives. And some of the things
that they're known to do to live this happier, healthier, more fulfilling life is eating a much healthier diet. Mostly plant based, but
not strictly vegetarian, practicing a healthier way
of life by sleeping well, staying physically active, and having alcohol
and moderation. And finally, having a sense
of purpose in their life. Many of them are active
in their communities, have a sense of duty to help their peers and practice
living minimally, which means they
don't waste time obsessing over needs
and material wants. But I want to know more
about how to live having. Now, where AI comes in on this is When you have this
subject in mind, you can ask AI to give you
even more recommendations. So from there, you can break it down where these blue zones are into the different parts of the world that these
blue zones are in. So then once you ask AI
about a certain topic, you're going to get answers
and ideas and suggestions. So then from there, what do
you need to consider next? You have to consider
what is worth sharing? What is an idea that
should be taught and what type of content that you can make
from this output. Now, if I asked
about blue zones, I'm going to get suggestions and ideas about the location, about their diet, and
about their lifestyle. From there, how can I break down these individual topics
into more specific topics? And this is where your expertise
has to come into play. If we're talking about location, what about these locations? Where does my expertise in mental health and
wellness know about how these certain locations can contribute to people's
overall well being. For example, if I'm
talking about the people who are in Okinawa that
lived by the ocean, then I can discuss
that location and the different parts of that place that
benefited these people. And then I would use AI
to augment my research. I would potentially even pull up research studies and
then link all of that into the AI and learn about what research
has been done there. And then from there, I can
create content on top of that. Now, as I was making
this video as well, I've noticed some of
the feedback that a lot of people were
asking about the diet, what people were eating. And so from there, I could propped onto Chat GBT or which AI
that I'm using how to improve my diet based on what these people were eating at the certain blue zone area
that we're referring to. And since there are
a multiple location, since there are many locations of blue zones around the world, We can focus on each individual
location separately. When we look at them separately, we can break down which part
of their diet works for them and then connect it to what you want to
teach your audience. Now, the goal of using Chat GBT for this is to
generate as much ideas as possible so that
you have a lot of different topics that
you can eventually cover. And as you have these topics, you could also use Chat GBT or any LLM to do some
more research for you? That way, you have
a list of topics. You have a list of
ideas on that topic, and then you can verify whether some of those
ideas make sense. And then you could also add
some more on top of that, maybe ask CGT to give you
different points of views or different ways to approach this if my target audience were
a certain group of people. You could say, like,
Hey, my certain audience is limited to here. Can you give me
some suggestions on a diet that is limited
to food in this region. So you need to be creative and
then with that creativity, tap into AI and
even more creative. That way, there's
an endless cycle of quality content because there is thought going into it and there's expertise
going into it as well, because you are the expert,
you're making content about a certain subject matter
that you have expertise in. Now, outside of this, outside of finishing
the step of using AI to supplement your research and your idea building is to
put your own spin on it. Where does your unique
approach come into play here? If, for example, you
are a business and you're a health and
wellness spa or something, you want to be able to
implement some of these into potentially an
article that you're writing or some advertisement
that you're doing, or maybe it's a yoga retreat, and you want to discuss some of the health and
wellness exercises that some of these
people are doing. Now, all of this is
just based off of the blue zones topic that
I mentioned in this video. And in this video, it really could have went so
many different directions because it was more of a general topic that
listed a lot of different locations
and lifestyles and some facts about it. So now that we know how
to make good content, which is really from starting
with your subject matter, expanding your subject matter, building some more research into that and then finally
putting your own spin to it. Up next, we're going to go in, and we're going to
create our own list of prompts as well and
to dive deeper into a certain subject so that
you can see what it's like as I am building content to talk about to
make videos for, to make blogs or to make
any type of content.
7. Prompt Formula for LLM's: Let's talk about
the prompt formula. So as you're writing something to communicate
with the AI, you need to have a formula or at least a
guideline that will help you get to a
better understanding of how to communicate
with the AI. And again, I'd like
to reiterate this is not artificial
intelligence yet. I know it's branded a lot as AI, artificial intelligence,
but in my personal opinion, we are far from there because
current AI technology is lacking a lot in the substance of what
it generates out. For example, as a
creative and as someone that's working
in video and film, if I ask AI to
generate me something, it's not going to generate all the little intricate details that will make a video or
film compelling and powerful. For example, I'm talking about sound design, custom design, the shots, the framing, the story, how it's all
implemented together, and how it's communicated, and then the vibe of
it, the color of it, and how all of that needs to be perfectly packaged to
have a powerful video. That's not going to
come about with AI yet. But using AI, you can study and break down
individual parts of that. So, for example, if you want to learn about
different angles, you can use AI to find out
about different angles. You can learn about
different color schemes and color palettes and you can learn about how different color
palettes go together. And then from all of this,
you can put it together to have a compelling piece
of creative work. But AI is not going to do it all for you just from
a single prompt. That's why we need
a prompt formula. That way you can
guide this AI to retrieve certain
information that it has in its knowledge base and then to output to
you so that you can have an answer that is suitable
for what your needs are. And as a quick rule of them, if your prompt is
extremely general, you're Output is also most likely going to be
extremely general as well, meaning if your prop is low
quality and not precise, then your output is also
going to be low quality. You can just go
in and say, write me a blog about so and so, because that's going to give
you very generic answers. And so as a
rudimentary guideline to formulating your prompt, we can consider a formula that
goes like this, the task, which is the desired output
that you want the AI to do, the context, which is what the AI should know about
what you're asking it, and then the parameters, which are basically terms and conditions that narrow down
your answer so that it is more fitting for you and not just something that the AI decides it wants you to spit out because it has basically infinite amounts of answers that it could give you. But if you are able to give it some guidelines on what
type of answer you want, then it's going to
significantly improve the quality of the answer
or output that you get. And some examples of
parameters could be like, give me answers in bullet points or give me this answer
in a certain tone, or maybe don't make stuff up, lower the temperature when
you are giving your response out So let's go through
some examples of prompts. I'll start with
something very simple, and we're going to
apply this formula to it so you can understand
what I'm talking about. The first one is summarize this transcript and give
it to me in bullet point. So the task for this is you
want the AI to summarize, and then you're going
to give it the context, which is a certain transcript,
maybe from YouTube, maybe from a video or maybe from a screenplay or even a blog. And then you want
your parameters to be in bullet points, as in, you want the answers to be in bullet points because
sometimes it might just give you a long paragraph or
multiple paragraphs and you can't read so much or you
don't want to read so much, so you ask it to give it
to you in bullet points. And maybe sometimes the bullet points are too much as well. So you could
reiterate on that and also add simplify
these bullet points. Now, this next
example, for the task, we're asking it to make
this title more fun. And then the title, the
context that we're giving it is learning AI for
content creation. And then finally, we
have the parameters. Make sure you include AI in this and give me a
couple examples. And then, of course, you can
get more specific by saying, have a similar word count, make this tone the same or keep the structure very
much similar to this. The more specific you
are in the constraints, the more precise the answer you're going to get
from the AI is. Now, this last one, which is
something I would use too, as I'm generating content
ideas for TikTok is tell me some more facts about
the Okinawan die Culture. So we just gave it the task, and then we gave it the context. And then finally,
for the parameters. I'm going to ask it
to give it to me in bullet points so that
it's summarized and then also include commonly
used ingredients or how they prepare this food
because maybe for my page, I'm doing a diet
page or I'm doing a health and wellness page
or a mental health page. So, in that case, I can ask
the answer to be more catered towards which specific scenario that I'm trying to
build content for. And so, before I
end this section, we need to discuss that a prompt is not going to give you the perfect
answer right away. Many times, it's going to need re iteration upon re iteration. So once you get an answer
out, you might think, Okay, this is not
fitting to what I want. So you have to go back and rewrite out and tell
it to do something else. Now, when you rewrite it again, you could also just
skip the first part, the context part and the
task part and just work on the parameters
because the output is a certain output already. So you're like,
Okay, this is too complicated. Let's
make it simpler. Let's make the sentence shorter. Or just give it to me
in three bullet points. And the goal of this is
for you to get closer and closer to the ideal
output that you want. And sometimes as
you're doing this, it might take you further
from the ideal output, so you have to
consistently or constantly reiterate on the prompt
that you're giving the AI. And sometimes if
it goes too far, you have to start
over because there's a certain limitation on
AI as of current times. So based on the
model you're using, maybe Chat GBT doesn't have the best memory recall or maybe Cloud doesn't
either or yet. So if it starts getting to Waki, you might have to go
back and restart it. But the thing is, there's
no perfect prompt. And the best way for you to get the answer that you
want from an AI is to constantly keep playing with this prompt or keep playing with the outputs until you
get something that you want. And a big tip, too, is the more time that you
spend on certain AI platforms, the better you're going to
get at communicating with these platforms because you wanted to retrieve information. Like we talked about
the Rag system, like how the whole
Rg system works. We wanted to retrieve information to give it to
us in a way that we want. So have to in a way with the
AI so that you can guide it to get this
information for you so that it can give you this information that
you're going to use. Otherwise, you're going to get low quality generic types
of output that is probably filling up the Internet
right now with a lot of low quality videos or low quality blogs,
and as a creator. And as someone that's
taking this class, I want you to have the
best quality content or best quality
creations out of AI. So in the next lecture, we're going to get
more in depth, and we're going to do
a hands on practice on how to work with LLMs, actually go in and check out these LLMs and see which
one is the best fit for you and how to negotiate around these large
language models.
8. Pratical Application (Hands-On Lesson): Alright, so this is the
fun video where we do a practical application and dive into a large
language model, and show you what goes on behind the scenes and how to
really augment your idea. Now, as a creative, your mission or your goal is to understand the different parts of what you're doing so that you can dive deep
into getting to your solution and
augmenting your ideas. So in this practice, I know that as I'm making
videos for my TikTok channel, which is in mental health and practical applications of it, that I have this clear goal
of informing my audience on ways to practice mental
health in a practical way. And so as you're breaking
down that process, for me to create content, I need to understand
what my audience wants, which is mental
health applications, how to dive deeper into certain topics about
mental health, and then how to dive even deeper into certain niches
in mental health. So in this video that made, they got over a couple of
hundred thousand views, it talks about different
groups of people. Now, in our practice, I'm going to choose
a certain group of people and then dive deep
into that group of people, understand why that group of people excel
in what they did, and then we're going to
tie it back to our page. And the takeaway from this is that you're giving
the AI a job to do, you're adding some
context to it, and then finally, you're
reframing by giving it these parameters for it to build some ideas for
you to work off of. And then finally,
you need to use your own experience
to add value to it. So before we jump into the
language model itself, I'm going to show you
this video so you get an understanding of what
the video is and then how we're going to use this video and break
down the process to expand on our ideas and to
have more content to make. The longest living
people in the world live in places considered blue zones. And these blue zones are
Okinawa, Japan, Sardinia, Italy, D Coya Costa Rica, Ikaria, Greece, and Low
Melinda, California. These people are known
to live well into their 90s and are happy
throughout most of their lives. And some of the things
that they're known to do to live this happier, healthier, more fulfilling life is eating a much healthier diet. Mostly plant based, but
not strictly vegetarian, practicing a healthier way
of life by sleeping well, staying physically active, and having alcohol
in moderation. And finally, having a sense
of purpose in their life. Many of them are active
in their communities, have a sense of duty to help their peers and practice
living minimally, which means they
don't waste time obsessing over needs
and material wants. Follow on to know
more about how to live Haby. Alright,
let's dive into it. So we're going to go
onto chat GPT now, and we're going to dive into our subject and explore what we can do with
their subject. Now, like I said before, we should know what your angle is as we are approaching this. So to do this, let's start off with something
super general. So I can show you what is
quality and what is not. So you could start
off by saying give me five topics to discuss
about blue zones. Now, it's going to
give you five topics. It's going to be very general. So here are two, let's
just go with response. One. And with these,
this is super general. Like, I could go any direction, but you need to understand where your audience is and what do
they come back to you for. So as I know that my
page is for wellness, mental health, and
practical applications, this is not going
to work for me. This is going to be something that is just very
general, right? Like, if I use this, it's
going to be very general. So I need to go a bit further. So then I would ask something about mental health potentially. So how can I apply
these to mental health? Give me some topics that
go in that direction. So give me some topics like mindfulness,
stress reduction, social connections,
purpose driven living, dietary influences. Now, of course, this is also very general.
So what do we do? We choose something in the
topic that we want to talk about and be extremely precise on where we
want to go with it. So I want to talk about the one of the topics in the blue zones that I was
mentioning in the video. So let's choose the
people of Okinawa. So I'm going to say, give me
five topics on the people of Okinawa in regards to their mental health practice,
what makes them happy. So this is more specific
towards our niche, which is mental health and what types of applicable
practices we can do. And also, we're focusing down on Okinawa because this video
showed some attraction, and Okinawa, people from Okinawa was one of the topics
in this video. So as we dive down here, we are introduced to a
couple of things here, like k guy, strong family ties, community engagement,
connection to nature, diet, and longevity. So from here, we
can actually start breaking it down and
go more complex. So here we start to learn about the concept
of maybe k guy, which is purpose in
life and how that influences happiness and is something that is
based on these people. So since our page is on
practical application, let's go ask about applications
of how to practices. So, I said, give me
five practices that we can do to practice CGI. So instead of practices, we can do practical
applications. And then as we are asking it to give us some practical
applications for this, it's giving these out, and we should tie back to the Okinawa people
so that this is at least more relevant to
what we've talked about. But if you were to go
and make this a video, you could already start
using this as a guideline on how to expand on this idea. But of course, I
want to tie it back to what we originally
talked about. So I would say
something like can we tie these back to
the Okinawan people. And then this is tied back to this group of people that we wanted to originally talk about, Here we have some
examples of what they do. So it says here, self reflection is
something that they do. Pursuing their passions is something that they're
doing as well, setting goals, community involvement
and career alignment. So from here, you have five different bullet points
for you to discuss. And as you are looking
at these bullet points, you need to start thinking about where your value
comes in as well. As a content creator, you have a lot of your
own personal knowledge. So from here, we can see how
the Okinawan people did it. So they did self
reflection as part of their community practices so that you need to start
thinking about how you do self reflection as your
own community practice. So as I'm going
through my channel, I'm bringing value by
providing my own insight, as well as reporting on what different things
people might not know, like the blue zones
and how people in Okinawa are part of these
blue zones. And so, For example, I could dig even deeper on self
reflection because I do a lot of journaling and journaling is a part
of self reflection. So I could go here and
even ask about what are some self reflection
practices that Okinawan people do that I can implement into my
journaling practice. So then from here, I
can expand and say, please give me at least
five journal prompts. And as we type this out, we got a few journal
prompts that would help connect this concept of self reflection
with mental health, to this group of people that
we're talking about that had shown or that had been shown
interest by your audience. So now we have a couple
of different elements tied together to create this
engaging piece of content. Now, the takeaway from
this is as you're writing your prompt and as you're reiterating and
building on that, you need to understand
what your end goal is. And as I'm creating videos, my end goal here
is to come up with practical applications
for someone to use in their daily life. And then based on
that topic that we discussed earlier about
the Okinawan people, I dug down into what their
life practices is like, and then we dug
deeper into doing some practical applications
like journaling. So from the journaling stuff, we have a couple of different
prompts here that we can use and we can share
about in our video. So then I could go off and use these five prompts
and discuss about it in my videos and
be like, Hey, today, we're going to talk about
the Okinawan people, and we are going to talk about one of the
things that they do a lot, and then we're going
to tie that back into some journaling practices
that you can take away today. And then I would list
these practices out, and then I would talk about it how I do these practices myself and how I can give examples of answering
these questions as well. But that could also
be another video. So you see here
that this expands your ideas out into many different directions
and dimensions.
9. Conclusion: Congratulations on
finishing the class. Now, I have to say you
really made it good. You made it all the
way to the end, so I have to congratulate
you because now you have the foundational
understanding of what AI is. And we now know that AI is a lot more than just
artificial intelligence, because now we learn that AI, how it works is really an information
retrieval system and that it's really here to
augment your intelligence. So, for me, AI is
augmented intelligence. And so for the project
that you're doing, show me how you're breaking down a topic, expanding on it. And in doing so, you
need to understand the thought process to get to
the solution that you want. So I'm very excited to see
what you guys come up with. Also, as a takeaway
from this class, you've now have the
foundational understanding of what AI is so that when you meet other people
and network with other people in the
AI, machine learning, deep learning space, you have this credibility so that when they talk to you and
when you talk to them, There's a certain
amount of respect because you understand what
is happening in the AI space. The space is
constantly evolving. There's going to be innovation month after month or
even week after week, sometimes even day after day. Stay up to date, and if you enjoy this class, please
leave me a review, let me know what
you like so that I can continue
doing more of that, and I'll see you
in the next video. Once again, congratulations.
Arnold Trinh, Mixed Media Creator
