AI and Chat GPT Deep Dive Part 1 – Transform Your Conversations and Ignite AI Success!!
In Part 1 - We look at the history of technology, different AI levels, what is Chat GPT and how you can better use it in your work and personal life.
(Title created by Chat GPT in the voice of Tony Robbins)
Amara’s Law:
“We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.”
Next week will be 6-months since the launch of Chat GPT, which has transformed and disrupted our world overnight.
Tech history has been paved by disruption and the content push for growth and expansion.
As Marc Andreessen stated back in 2011, software is eating the world.
While the last 40 years have been amazing for technology, we are just now entering the exponential phase of that growth through the introduction of Artificial Intelligence to the masses.
Quick History of Technology Last 40 Years
We first faced the introduction of hardware to the masses through computers, printers, and mouses.
What is hardware? Let Chat GPT answer that for you in a 3rd-grade rhyme.
Once people had computers in their homes, they really had no idea how to use them.
These new things were cumbersome, very technical, and required “skill” to manipulate. If anyone ever used DOS, you know what I am talking about.
Then came along this dude named Bill Gates who streamlined the “software” of MS-DOS to make it easy to use.
With a click of the button, a window would pop up and you can just click what you want to use.
Windows was really the introduction of “software” to the masses to enable them to use hardware better.
This is what created the goliaths we call Microsoft and Apple.
After the introduction of software to the masses, people wanted to start to talk to each other.
This created the internet.
But once everyone had AOL and signed in every day to listen to this sound…
Before the magic words came up saying…..
We all started to create.
We created emails, then webpages, then pictures, and videos. The amount of data we started to create skyrocketed higher.
All this data created the next major hurdle. Where to store it all.
“More data is created per hour today than in an entire year just two decades ago, and global data is expected to double every 2 years.”
– Bank of America
Storage was and continues to be a major issue.
This started the major storage wars driven by Western Digital, Seagate, Micron Technologies, Samsung, and others.
“More data is created every hour today than in an entire year just 20 years ago.”
– Seagate Technologies
But if you were a computer owner back in the early to mid-2000s, you probably always ran into storage issues.
On my first iPhone, storage always maxed out and I had to delete photos constantly.
Besides storage, the other major issue was getting the software to you more efficiently.
AOl back in the late 1990s sent the disks above to everyone in the mail.
Over 50% of all the CD disks worldwide at that time had the AOL logo on them.
It is said AOL spent over $300 MILLION a year just on US postage to get these CDs to your home. And by the time everyone received them, there was a new version of the software that fixed bugs and ran better, so it had to start over again.
There had to be an easier way.
Then this magical thing called “the cloud” came along to solve our storage and software issues.
Instead of storing all your data yourself, you can just upload that data to a data server or farm, for a small fee, and you can create as much data as you wanted.
In a business model framework, we went from a point-to-point model, pushing everything out to each customer, to a hub and spoke model, where the information and services were in one hub and there was communication back and forth.
A new version of software coming out. FIne, we will push it out to everyone through the cloud while they sleep.
Are YOu having technical issues with your software? Fine, we will fix it through the cloud.
This created a massive amount of efficiency in technology, making it a win-win-win all around for the provider, consumer, and service side.
This idea was captured by Microsoft, Amazon AWS, Apple, Google, and many others and was the driver of most of the technology sector growth over the past 10 years.
Once cloud computing was introduced and we no longer had storage issues, the amount of storage created escalated significantly.
“We will generate more data in the next 2 days than all the data created between the dawn of humanity and 2000.”
– Eric Schmidt
“Today it would take a person approximately 181 million years to download all the data from the internet.”
- Unicorn Insights
We now create a massive amount of data each and every minute of the day.
This data created our next major problem. How do we figure out how to use all of it?
My Evernote account is a perfect example.
I opened this account in 2006 at the birth of the cloud idea to capture all my notes from work and personal life. I currently have over 15,000 individual notes and find it extremely difficult to pull and use any old data.
Forget about connecting the dots. Using data from 8 years ago with something similar that I found 2 years ago. It is nearly impossible.
We are failing miserably at being able to use, manipulate, and implement data.
Humans do a very bad job of connecting the dots.
As of today, only 1% of all global data is being captured, stored, and USED.
This is where AI kicks in.
What is Artificial Intelligence
Since the intro of the computer in the 1950s, till 6 months ago, software was considered very dumb.
As stated in Wire magazine:
“Most enterprise software simply serves as a tool to get a particular job done faster or more efficiently, but few applications are really “smart”.”
https://www.wired.com/insights/2014/04/end-dumb-software/
To call software smart, it needs to leverage the data we have stored and come up with predictions and analytics that work to spot patterns.
That is what AI specializes in.
As the founder of Chat GPT stated, AI is just “smart software”.
Professor Yutaka Matsuo divided AI into 4 unique levels. (below is data provided by Bank of America research team).
Level 1 – Simple Control Programs
This is a level of software mainly for marketing purposes. Think of air conditioner systems, cleaners, and washing machines. That special setting on those machines. Ya, that is level 1 controls.
Level 2- Classic Artificial Intelligence
Artificial intelligence is capable of responding to questions. Think of puzzle-solving programs and diagnostic programs. The key here is this level is only as smart as the amount of data or knowledge you input into it.
The perfect example of level 2 AI is in your pocket. The Apple Siri bot or Amazon’s Alexa is level 2.
It can understand and respond to voice commands and questions and do simple tasks. But I think we would all agree as my mom would say.
“Alexa is just dumb. She does not understand me and what I want most of the time.”
-My 87 Year Old Mom
Level 3- Artificial Intelligence embodying machine learning
This is what is embedded in search engines and can make decisions on big data. This is where machine learning comes in.
The computer itself can extract rules and decision-making standards based on the data without those rules being explicitly given in the data. In essence, the machine can write new programs instead of humans. Increasing the amount of data will result in an improvement in the accuracy of output.
At level 3, the machine can connect the dots. You will no longer think it’s stupid or dumb. It will blow you away with how accurate and human-like it really is.
From Bank of America’s research team:
This is where Chat GPT sits.
As Chat GPT told me this morning, level 3 is its classification.
Chat GPT is at the level of what was framed in the famous movie “Her”, which came out 9 years ago.
Besides the productivity changes, education, and other components Chat GPT can help with (I highlight below), many experts feel kids and young adults will have an AI bot they can talk to like the movie Her.
If we stopped today at level 3, the world as we know it today will be dramatically different in 5-10 years.
What we have seen over the last 40 years is nothing compared to what is coming.
But there is still a next level. The point of no return. Level 4.
Level 4- Artificial Intelligence capable of independent program generation
Here computers can learn the feature values necessary for displaying the data used for machine learning.
This is where deep learning kicks in. Deep learning is machine learning that uses neural networks like our brain.
What separates this level from others is the ability to extract information and then generate new programs or ideas from that information. To learn. To create. Just like a human.
I asked Chat GPT what needs to happen for it to reach level 4. Below is its answer.
I think the movie Ex Machina, which came out 8 years ago, is a perfect example of what level 4 AI can actually do when perfected.
This is the level all experts are scared of. This is the level of no return.
What is Chat GPT
I asked Chat GPT to give you the four steps above in a rhyme so a 3rd grader can understand it.
Here is its reply.
What is Chat GPT? Let’s have it tell you.
Sure, here's an explanation of what I am, in rhyme for a third grader:
I'm an AI, here to help and play,
I can answer questions, come what may.
Just type your thoughts, and I'll reply, With information, I'll give it a try!
I know about many things, far and wide,
From animals, planets, to the ocean's tide.
I'll find facts and figures, all neat and clear, To make learning fun, that's why I'm here!
But remember, dear friend, I'm just a machine, No emotions or thoughts, that's the scene.
I can chat and converse, but in the end, It's you who brings the heart and mind blend!
So ask away, my curious mate,
Together, let's explore and create.
We'll learn and grow, side by side,
With knowledge and fun, let's enjoy the ride!
Large language models like Chat GPT are trained on vast amounts of data, learning this data, and then understanding and generating human-like text messages from this data.
Below is a TED Talk on the power of this new technology.
If you do not want to watch all 30 minutes, then go to 1 minute 30 seconds in and watch where he made it come up with a dinner plan for the night, and then use the app plug-in to automatically order the food through Instacart.
“The intensity of just one day's worth of AI training is equal to 150 years of gaming training for a human.”
– Bank of America data from Tencent
The GPT in Chat GPT stands for Generative Pre-trained Transformer.
Its goal is to generate human-like text from a pre-trained large data set. All it does is take data and aggregate it the best it can to give you a “smart” response.
Unlike Google search, where you look up something and then Google internet algorithms go out into the internet to find your answer, Chat GPT generates its own answer.
To find something on Google, someone else needed to type it, ask for it, or video it in the past. Chat GPT, in theory, gives you brand-new never seen responses.
Unlike the years it took to go from basic software design and delivery, Chat GPT has been designed from the bottom up to be user-friendly.
It is something a 3-year-old to a 105-year-old can easily use.
As a result, its rapid adoption into society is like something we have never seen before.
It took Netflix 3.5 years to reach 1 million users. It took Facebook 2.5 years to reach 1 million users. Twitter took 2 years.
Chat GPT took less than 2 weeks.
How to use Chat GPT
Just like the introduction of the computer, we now have this all-powerful thing that many have no idea how to use.
We need to teach ourselves how to use this new technology. Many are using it like Google, which is wrong.
All responses from Chat GPT are original whereas Google just uses previously looked-up terms, phrases, and ideas that others have looked up before and points you in that direction.
I have been spending a lot of time trying to optimize my uses of Chat GPT. Frankly, I still have a very long way to go.
But I wanted to highlight a few things I have come across that have helped me.
The first thing you will notice about Chat GPT is that it is very long-winded sometimes.
The designers have given it so many guard rails that sometimes you have to read a ton of crap just to get to the real answer.
So the below Reddit user had a good idea to get around this.
Here is another more detailed prompt.
But once we get past this small step, what can you actually use this for?
Well, for one, Chat GPT can be a fantastic tutor. I really think within the next 5-10 years, this software will forever change our education system.
When I was a kid I hated reading. I think I read 1 book in my life up to my senior year in college. But that year I moved to Australia and lived with a kid who always read.
So one day he took me to a large bookstore and said, I know you are a person who loves to learn. Think about this bookstore this way; There is a book on every topic you can think of and someone has already faced and written about all the problems and hurdles they had to go through to get to the finish line.
Find what you are interested in and go browse the books on that topic.
At the time I wanted to know about money and some futuristic science stuff.
That day I purchased three books i still have today:
Rich Dad, Poor Dad,
The Intelligent Investor, and
Chaos Theory, the new Science.
By the time I got to page 30 in Rich Dad, Poor Dad, I was hooked.
I now have over 800 books on my personal Kindle and read them daily.
Why am I telling you this?
Because the school system as we know it failed me. The forcing of subjects on me I did not care about and the one size fits all process of the education system failed to get me excited about anything.
Our education system was designed during the industrial revolution. It was made to give a base level of education to the masses. It was designed for yesterday’s work.
I promise you this. The future is Chat GPT and other language bots. No longer do we need teachers to teach. You have a personal assistant and tutor in your pocket who can teach you anything you want. Then the teacher can be the hands-on person to drive the points home. Everyone can learn at their own pace. Everyone can be taught in the best way that fits their unique personality.
Just look at the below post from a Reddit user.
Chat GPT will separate the haves and have-nots more as we move into the future.
But it will not be by race or income level or status level. It will be on a curiosity level.
If you are curious enough, the sky is the limit with this new technology.
The good news is most babies and toddlers are naturally curious. I also think our current education system, with the rules and tests and sitting all day, beats out this naturally curious level of kids.
I really think this technology will continue this natural born trait to levels we cannot even imagine.
We all just need to be like George again.
Besides teaching you things, it can help you work.
But to be able to do this, we need to write prompts.
Right now, no one really knows how to write the best prompt.
The below post can help with this. Instead of guessing what will give us the best answer, we ask Chat GPT to help us write our prompts.
I used the above to help me write a Family Trust. If done right, you will start to see the real magic of this technology.
Finally, I wanted to share a tweetstorm with you by someone who uses this for his marketing and writing business.
I think he outlines nicely the limitations of this technology and what we need to do, as the users, to make it help us better.
As a writer, I find the below fascinating.
This is just a framework for writing.
But this type of framework can be created for any and all professions that think and create.
Just remember, Chat GPT is as good as the prompt you give it.
If you want to learn more about this, Stanford did a free online learning webinar on this topic. The link is below.
https://online.stanford.edu/events/202305/how-you-can-use-chatgpt-increase-your-creative-output
I hope you learned something from this today. I would love to hear back on your thoughts.
Have a wonderful long holiday weekend.
Next week we will get back to the market and economic stuff.