OpenAI.com - ChatGPT

1ThoughtKnown

I’ve been spending time on ChatGPT, engaging in conversations with the AI. The future of AI is now and the potential is incredible. It can be quite entertaining. Asked it to write a poem for me:

1ThoughtKnown

Sorry that the photos are different sizes. But you get the point 👍

1ThoughtKnown

Thought I would carry on and ask for another. I carried on quite the conversation with the AI about PJ after. According to the AI, PJ has sold 85 million albums as of September 2021 when the information was provided. The AI does not get its info from the internet, thus eliminating misinformation:

Halifax2TheMax

Now ask it to write a poem about Kiss.

1ThoughtKnown

Last one. Remarkable what this thing generates in 15 seconds:

1ThoughtKnown

Halifax2TheMax said:

Now ask it to write a poem about Kiss.

I’ll leave that for Kiss fans, as I am not one of them. 

Hi!

^Luv the hiking one. 

I hike these mountains for my soul
I hike these mountains to feel whole

Not bad. At all.

1ThoughtKnown

Hi! said:

^Luv the hiking one. 

I hike these mountains for my soul
I hike these mountains to feel whole

Not bad. At all.

Agreed! This is all rather frivolous of course, but I was having conversations with the AI about political ideologies, religious views, philosophers and their importance like Kants and Locke; in addition to having it write poems and lyrics. Also had it construct for me executive summaries on a variety of topics, short stories, movie plots. It’s mind boggling. 

Hi!

^I’ve messed around with some AI generated things before for funnzies but nothing on the level of which you’re speaking of. Mainly just fun song titles about Trump and golden toilets. Lol Some of the AI generated art I’ve seen is intriguing. Will delve more.🤙🏻

brianlux

It all sounds like harmless fun but I'm one of those skeptical people who believe we are being lulled deeper in an artificial world.  Another reason why I take walks in the outdoors as often as possible- 359 days last calendar year alone.

josevolution

brianlux said:

It all sounds like harmless fun but I'm one of those skeptical people who believe we are being lulled deeper in an artificial world.  Another reason why I take walks in the outdoors as often as possible- 359 days last calendar year alone.

Yeah that’s why I’m down here in Chile this week heading to Patagonia for a walk on the wild side 👍 there’s nothing artificial except the wooden huts I’ll be sleeping in

benjs

brianlux said:

It all sounds like harmless fun but I'm one of those skeptical people who believe we are being lulled deeper in an artificial world.  Another reason why I take walks in the outdoors as often as possible- 359 days last calendar year alone.

There are an abundance of pros and cons to AI (which I'll call Machine Learning from hereon, since AI is a misnomer). For each positive, I can imagine a large field of spiky pits if not leveraged properly/fairly. 

On ML for art - I love it, and see it as a tool similar to Photoshop, but without learning how to 'ask' the system for changes (aka learning Photoshop). Check out this link - https://andys.page/posts/how-to-draw/. This artist's usage of Stable Diffusion (a series of image-drawing machine learning tools by those who make ChatGPT), really suggests to me that ML can be used in artistic ways rather than purely robotic, so long as the knowledge seed is human. Specifically, in this venture, ML could not replace a human to an interesting artistic end in the linked case, but could aid one effectively. If the human pursuit is ML as an empowering extension of the human brain (rather than a replacement) for artistic ventures, I can see this as a net positive. The major caveat and concern is a licensing model for art generated and used by ML models for training purposes, so that all prior works are rewarded (think the digital music industry licensing concerns, but add complexity). 

On ML for programming - I'm a huge fan. Good programming = accurately conveyed software goals and requirements from a client, practical organization of code, proper ability to produce conveyed goals and requirements, a robust error- or typo- checking procedure, production of pseudo-code in the form of code comments to explain what the code's intended to do - all produced leveraging known and documented design patterns for optimal code. Not only can ML assist with performing any of the above steps, but it can also reverse engineer how it would have done an above step - serving as a 'code check' assistant to help find errors or scope that still needs to be coded. When ML reaches sufficient maturity (probably no more than a couple of years), development of internal tools or external products need not be exclusively for/by coders, and the business user's toolkit will expand to include development tools that are 'talked to' instead of writing code. This can be an empowerment of an existing labour force, or can bring in adjacent labourers (businesspeople) to aid in developmental assistance. Sure, it can be used to reduce cost centres, but at the expense of quality of output. No businesses will lay off due to ML if their competitors are enhancing their businesses through ML instead. That would add a further gap between those two types of businesses (progressive vs. frugal), disadvantaging the latter seriously.

On ML for hand-work (instead of creative work or logic-creating work - which I'll call robotics) - this one seriously worries me. Too vast a chunk of the population makes a living off of pure hand-work, that's too easily replicated with machinery that only costs an initial investment plus maintenance costs along the way - compared to an employer paying for an ongoing salary and all that comes with it (human variability, requests for vacation, sick days, etc.). I personally can't think of a way for machinery to aid operators instead of replacing them without a net profitability decrease. Conversely, most replacement strategies in this realm could lead to a realistic increase in profit. I don't see the value proposition to a businessperson to hold onto extra human labourers instead of acquiring robots, when this becomes an option in their realm. 

The long and short of what I'm trying to say - is that ML is a tool, and we should bring it into the world with organization and intentionality. Just as we scoff at mathematicians who defer to an abacus instead of a calculator today, we'll scoff at those who try to code, draw, optimize businesses without the assistance of ML, and consider them outdated and likely lacking compared to their competition. 

It's already late to train the world on how ML can/will change the world - governments and education programs need to embrace it ASAP lest it brings further wealth inequality to already dangerous levels.

benjs

josevolution said:

brianlux said:

It all sounds like harmless fun but I'm one of those skeptical people who believe we are being lulled deeper in an artificial world.  Another reason why I take walks in the outdoors as often as possible- 359 days last calendar year alone.

Yeah that’s why I’m down here in Chile this week heading to Patagonia for a walk on the wild side 👍 there’s nothing artificial except the wooden huts I’ll be sleeping in

Amazing, Jose! Have an amazing time. I had planned to go two years ago, but, y'know, CoVID.

static111

It’s all fun and games until you ask to solve pi and it destroys the universe in an attempt to gather more computing power.

josevolution

benjs said:

josevolution said:

brianlux said:

It all sounds like harmless fun but I'm one of those skeptical people who believe we are being lulled deeper in an artificial world.  Another reason why I take walks in the outdoors as often as possible- 359 days last calendar year alone.

Yeah that’s why I’m down here in Chile this week heading to Patagonia for a walk on the wild side 👍 there’s nothing artificial except the wooden huts I’ll be sleeping in

Amazing, Jose! Have an amazing time. I had planned to go two years ago, but, y'know, CoVID.

Whoa yeah me two I had to cancel it probably was around the same time! Chile is amazing 🤩 

1ThoughtKnown

brianlux said:

It all sounds like harmless fun but I'm one of those skeptical people who believe we are being lulled deeper in an artificial world.  Another reason why I take walks in the outdoors as often as possible- 359 days last calendar year alone.

I do that quite often myself, but playing around with technology is also fun. There is some professional advantage to getting out in front of stuff if I can use it to my advantage. There is plenty of good information available on there. 

1ThoughtKnown

As far as benjs comment about what it is called, my nephew works in AI, he doesn’t call it Machine Learning, so it’s hardly a misnomer. As a matter of fact, he had dinner with the person running OpenAI a couple months ago. 

Anyways, I decided to ask the expert on what the proper term is:
“ Are you machine learning or AI?

I am an AI assistant trained to assist users with a variety of tasks, including answering questions and providing information on a wide range of topics. My training involved the use of machine learning techniques, which allowed me to learn and improve my performance over time by analyzing large amounts of data.”

1ThoughtKnown

Then I asked:

So what is machine learning?

Machine learning is a subfield of artificial intelligence that involves the use of algorithms and statistical models to enable computers to learn and make decisions with minimal human intervention. It involves training algorithms using large sets of data, which can be used to make predictions or take actions based on new input.

brianlux

josevolution said:

brianlux said:

It all sounds like harmless fun but I'm one of those skeptical people who believe we are being lulled deeper in an artificial world.  Another reason why I take walks in the outdoors as often as possible- 359 days last calendar year alone.

Yeah that’s why I’m down here in Chile this week heading to Patagonia for a walk on the wild side 👍 there’s nothing artificial except the wooden huts I’ll be sleeping in

That sounds fantastic, Jose!  Have a great time!

benjs said:

brianlux said:

It all sounds like harmless fun but I'm one of those skeptical people who believe we are being lulled deeper in an artificial world.  Another reason why I take walks in the outdoors as often as possible- 359 days last calendar year alone.

There are an abundance of pros and cons to AI (which I'll call Machine Learning from hereon, since AI is a misnomer). For each positive, I can imagine a large field of spiky pits if not leveraged properly/fairly. 

On ML for art - I love it, and see it as a tool similar to Photoshop, but without learning how to 'ask' the system for changes (aka learning Photoshop). Check out this link - https://andys.page/posts/how-to-draw/. This artist's usage of Stable Diffusion (a series of image-drawing machine learning tools by those who make ChatGPT), really suggests to me that ML can be used in artistic ways rather than purely robotic, so long as the knowledge seed is human. Specifically, in this venture, ML could not replace a human to an interesting artistic end in the linked case, but could aid one effectively. If the human pursuit is ML as an empowering extension of the human brain (rather than a replacement) for artistic ventures, I can see this as a net positive. The major caveat and concern is a licensing model for art generated and used by ML models for training purposes, so that all prior works are rewarded (think the digital music industry licensing concerns, but add complexity). 

On ML for programming - I'm a huge fan. Good programming = accurately conveyed software goals and requirements from a client, practical organization of code, proper ability to produce conveyed goals and requirements, a robust error- or typo- checking procedure, production of pseudo-code in the form of code comments to explain what the code's intended to do - all produced leveraging known and documented design patterns for optimal code. Not only can ML assist with performing any of the above steps, but it can also reverse engineer how it would have done an above step - serving as a 'code check' assistant to help find errors or scope that still needs to be coded. When ML reaches sufficient maturity (probably no more than a couple of years), development of internal tools or external products need not be exclusively for/by coders, and the business user's toolkit will expand to include development tools that are 'talked to' instead of writing code. This can be an empowerment of an existing labour force, or can bring in adjacent labourers (businesspeople) to aid in developmental assistance. Sure, it can be used to reduce cost centres, but at the expense of quality of output. No businesses will lay off due to ML if their competitors are enhancing their businesses through ML instead. That would add a further gap between those two types of businesses (progressive vs. frugal), disadvantaging the latter seriously.

On ML for hand-work (instead of creative work or logic-creating work - which I'll call robotics) - this one seriously worries me. Too vast a chunk of the population makes a living off of pure hand-work, that's too easily replicated with machinery that only costs an initial investment plus maintenance costs along the way - compared to an employer paying for an ongoing salary and all that comes with it (human variability, requests for vacation, sick days, etc.). I personally can't think of a way for machinery to aid operators instead of replacing them without a net profitability decrease. Conversely, most replacement strategies in this realm could lead to a realistic increase in profit. I don't see the value proposition to a businessperson to hold onto extra human labourers instead of acquiring robots, when this becomes an option in their realm. 

The long and short of what I'm trying to say - is that ML is a tool, and we should bring it into the world with organization and intentionality. Just as we scoff at mathematicians who defer to an abacus instead of a calculator today, we'll scoff at those who try to code, draw, optimize businesses without the assistance of ML, and consider them outdated and likely lacking compared to their competition. 

It's already late to train the world on how ML can/will change the world - governments and education programs need to embrace it ASAP lest it brings further wealth inequality to already dangerous levels.

Interesting stuff there, Ben. 

I guess being a fan of Neil Young's album Trans or Kraftwerk's Autobahn would mean I'm at least somewhat accepting of ML in the arts. I'm always going to go back to "flawed" human produced music as my go-to, but I do appreciate some creative aspects that a tech dependent.

As for robotics replacing hand labor, it's mostly too late to slow that down.  Both can exist of course, but hand labor is becoming more a hobby type of activity.  Look at knitting, for example.  There are machines that mass-produce clothing, knitting machines that hobbyist use at home to make things, and then people like my wife, C., who just finished hand knitting a sweater.  That latter accomplishment amazes me the most yet that's how they were all made at one time.

In any case, yes, ML as a tool is here to stay and I do hope it can be used more for the good, less for inequality.

1ThoughtKnown said:

brianlux said:

It all sounds like harmless fun but I'm one of those skeptical people who believe we are being lulled deeper in an artificial world.  Another reason why I take walks in the outdoors as often as possible- 359 days last calendar year alone.

I do that quite often myself, but playing around with technology is also fun. There is some professional advantage to getting out in front of stuff if I can use it to my advantage. There is plenty of good information available on there. 

I probably do deserve to be regarded as a "Dinosaur".   

1ThoughtKnown

brianlux said:

josevolution said:

brianlux said:

It all sounds like harmless fun but I'm one of those skeptical people who believe we are being lulled deeper in an artificial world.  Another reason why I take walks in the outdoors as often as possible- 359 days last calendar year alone.

Yeah that’s why I’m down here in Chile this week heading to Patagonia for a walk on the wild side 👍 there’s nothing artificial except the wooden huts I’ll be sleeping in

That sounds fantastic, Jose!  Have a great time!

benjs said:

brianlux said:

It all sounds like harmless fun but I'm one of those skeptical people who believe we are being lulled deeper in an artificial world.  Another reason why I take walks in the outdoors as often as possible- 359 days last calendar year alone.

There are an abundance of pros and cons to AI (which I'll call Machine Learning from hereon, since AI is a misnomer). For each positive, I can imagine a large field of spiky pits if not leveraged properly/fairly. 

On ML for art - I love it, and see it as a tool similar to Photoshop, but without learning how to 'ask' the system for changes (aka learning Photoshop). Check out this link - https://andys.page/posts/how-to-draw/. This artist's usage of Stable Diffusion (a series of image-drawing machine learning tools by those who make ChatGPT), really suggests to me that ML can be used in artistic ways rather than purely robotic, so long as the knowledge seed is human. Specifically, in this venture, ML could not replace a human to an interesting artistic end in the linked case, but could aid one effectively. If the human pursuit is ML as an empowering extension of the human brain (rather than a replacement) for artistic ventures, I can see this as a net positive. The major caveat and concern is a licensing model for art generated and used by ML models for training purposes, so that all prior works are rewarded (think the digital music industry licensing concerns, but add complexity). 

On ML for programming - I'm a huge fan. Good programming = accurately conveyed software goals and requirements from a client, practical organization of code, proper ability to produce conveyed goals and requirements, a robust error- or typo- checking procedure, production of pseudo-code in the form of code comments to explain what the code's intended to do - all produced leveraging known and documented design patterns for optimal code. Not only can ML assist with performing any of the above steps, but it can also reverse engineer how it would have done an above step - serving as a 'code check' assistant to help find errors or scope that still needs to be coded. When ML reaches sufficient maturity (probably no more than a couple of years), development of internal tools or external products need not be exclusively for/by coders, and the business user's toolkit will expand to include development tools that are 'talked to' instead of writing code. This can be an empowerment of an existing labour force, or can bring in adjacent labourers (businesspeople) to aid in developmental assistance. Sure, it can be used to reduce cost centres, but at the expense of quality of output. No businesses will lay off due to ML if their competitors are enhancing their businesses through ML instead. That would add a further gap between those two types of businesses (progressive vs. frugal), disadvantaging the latter seriously.

On ML for hand-work (instead of creative work or logic-creating work - which I'll call robotics) - this one seriously worries me. Too vast a chunk of the population makes a living off of pure hand-work, that's too easily replicated with machinery that only costs an initial investment plus maintenance costs along the way - compared to an employer paying for an ongoing salary and all that comes with it (human variability, requests for vacation, sick days, etc.). I personally can't think of a way for machinery to aid operators instead of replacing them without a net profitability decrease. Conversely, most replacement strategies in this realm could lead to a realistic increase in profit. I don't see the value proposition to a businessperson to hold onto extra human labourers instead of acquiring robots, when this becomes an option in their realm. 

The long and short of what I'm trying to say - is that ML is a tool, and we should bring it into the world with organization and intentionality. Just as we scoff at mathematicians who defer to an abacus instead of a calculator today, we'll scoff at those who try to code, draw, optimize businesses without the assistance of ML, and consider them outdated and likely lacking compared to their competition. 

It's already late to train the world on how ML can/will change the world - governments and education programs need to embrace it ASAP lest it brings further wealth inequality to already dangerous levels.

Interesting stuff there, Ben. 

I guess being a fan of Neil Young's album Trans or Kraftwerk's Autobahn would mean I'm at least somewhat accepting of ML in the arts. I'm always going to go back to "flawed" human produced music as my go-to, but I do appreciate some creative aspects that a tech dependent.

As for robotics replacing hand labor, it's mostly too late to slow that down.  Both can exist of course, but hand labor is becoming more a hobby type of activity.  Look at knitting, for example.  There are machines that mass-produce clothing, knitting machines that hobbyist use at home to make things, and then people like my wife, C., who just finished hand knitting a sweater.  That latter accomplishment amazes me the most yet that's how they were all made at one time.

In any case, yes, ML as a tool is here to stay and I do hope it can be used more for the good, less for inequality.

1ThoughtKnown said:

brianlux said:

It all sounds like harmless fun but I'm one of those skeptical people who believe we are being lulled deeper in an artificial world.  Another reason why I take walks in the outdoors as often as possible- 359 days last calendar year alone.

I do that quite often myself, but playing around with technology is also fun. There is some professional advantage to getting out in front of stuff if I can use it to my advantage. There is plenty of good information available on there. 

I probably do deserve to be regarded as a "Dinosaur".   

I look at the work front completely differently. As a Health and Safety Professional, I look at robotics and AI as a way to save lives by allowing the machinery to perform high risk tasks so people don’t. If you take human error out of the equation, “using their hands” so-to-speak, you greatly reduce risk in workplaces. It is good business. If you don’t like spending copious amounts of time mining raw data, then AI is your friend. Couldn’t those people be put to better use? Sitting in front of a desk for 8-10 hours a day… the long term health effects due to poor ergonomics are immense. 

The possibilities are endless. 
 

The fear that humans will always have for any new technology will always exist. The vacuum of jobs that won’t exist anymore and ethical questions will always be there, but there are plenty of options. Im sure the western world will create more jobs as the technology becomes more prevalent, jobs we just couldn’t imagine. In 2008, did you see companies hiring executive team members for social media? I sure didn’t. 

There is experiments done with Guaranteed Basic Income. One done in Canada long ago was quite successful, yet the idea was abandoned for unknown reasons, probably because it may have eliminated several government social programs and eliminated several jobs in the process. 

My nephew called this his generations moon landing. He is over the moon with excitement (pun intended). He built a self driving car at age 15 and has followed his dreams.