A Conversation with Chat GPT

Chilean Chit-Chat
1

Chat GPT - a language model / artificial intelligence/ chat bot - has been on the news lately, usually when it says something dumb or offensive, but those are the minority of cases. The majority of the time it is impressive. Here is a conversation I had with it:

Is resistance to change a heritable trait? Is openness and willingness to try new things a heritable trait? Is the disgust reaction a heritable trait? Please provide either sources or evidence for your claims. Also please suggest where on the internet or in books I might be available to find more information about this?

Resistance to change, openness and willingness to try new things, and the disgust reaction are all traits that have been studied in relation to genetics and heritability.

Research has shown that some aspects of personality, such as resistance to change and openness to new experiences, have a moderate to high heritability, meaning that genetics plays a significant role in determining these traits. However, it is important to note that heritability does not mean that these traits are solely determined by genetics, and environmental factors also play a role.

The disgust reaction, which is related to the immune system and is thought to have evolved as a way to protect against pathogens, has also been studied in relation to genetics. Studies have found that certain genetic variations are associated with increased or decreased disgust sensitivity, which suggests a heritable component to this trait.

For more information about heritability and personality, you might try looking for articles and studies in academic journals such as “Personality and Social Psychology Review” and “Journal of Personality”. Additionally, books such as “Personality: What Makes You the Way You Are” by David C. Funder and “The Handbook of Evolutionary Psychology” edited by David M. Buss might be useful resources.

Let me tell you my theory and then perhaps you can critique it and tell me how likely you think it is true. My theory is that people with higher resistance to change would have had a higher chance of survival thousands of years ago because they would have stayed in safe places and only eaten foods known to be safe. People with higher willingness to try things might have explored more and therefore been killed by predators or rival groups, or tried different foods more willingly and then been poisoned and died. Given the heritability of these traits, it may be that we humans still today have the evolutionary psychology that was useful thousands of years ago, but is less optimal today. This would cause humanity to be overly resistance to change today, for example in relation to replacing fossil fuels with low carbon energy. What do you think?

The theory you’ve described about the evolution of resistance to change and willingness to try new things is an interesting one, and it’s certainly possible that the traits you’ve described could have been selected for in the past. However, it’s important to note that evolution is a complex process and the factors that influence the selection of certain traits can vary depending on the specific context. Additionally, while there may be some genetic basis to these traits, they are also likely influenced by environmental factors and individual experiences.

As for the specific example of resistance to change in relation to replacing fossil fuels with low-carbon energy, it’s certainly true that there are many factors that can make it difficult for individuals and societies to change their behaviors and adopt new technologies. However, it’s also worth considering that there are many other factors that can influence this process, such as economic, political, and social considerations.

It’s also worth noting that the heritability of psychological traits is a complex and controversial topic and more research is needed to understand the specific genetic and environmental factors that contribute to these traits.

You may want to read some books on Evolutionary psychology, and also research articles on heritability of behavioral traits and the influence of culture, society, and environment on behavior change.

1 Like
2

I have read about Chat GPT and so created an account to chat a week ago. I have asked to write a poem about a cat. Then about two cats. The last try was to write a poem about one thousand and thirty three cats. Then about a tank and a missile (in the poem the missile hits the tank). Then about a missile and a fighter jet. (This time the jet escaped). I asked to write a poem about various subjects and the response was way more interesting than when asking direct questions. The length of these poems was about 16 to 20 lines. Also I asked to write haiku about various topics.

2 Likes
3

compose a poem about a crocodile, a soldier, a mouse and a nuclear bomb:

A crocodile, a soldier, a mouse, a bomb,
A curious cast, all gathered in one.
The swampy banks hold the first, with a grin,
A soldier, alert, stands watch from within.

A tiny mouse, skittering by with a squeak,
Seems an unlikely part of this unique
Assembly, but scurrying, it takes its place,
In the shadow of the great, fearsome crocodile’s face.

The final piece of this strange, eerie scene,
Is a nuclear bomb, that gleams like a dream.
Its destructive power, a warning untold,
A weapon of mass destruction, uncontrolled.

This unlikely quartet, together they stand,
In a world where peace is fragile and bland.
A symbol of all that’s wrong, that we see,
In a world where danger surrounds, constantly.

So let us remember this curious crew,
And all they represent, so stark and so true.
For they show us that even the small, and meek,
Can play their part in the dangers we seek.

4

That’s very good in my opinion considering it was a hard task given. I can write fairly well but I couldn’t beat that.

5

A crocodile, a soldier, a mouse, a bomb,
A curious cast I think you’ll agree.
Unless they’ve been assembled to test Me.

Unlike the machine,
I see the Why,
And refuse to comply

With an added coda…

So Eff Wye…

6

Chat Haiku (with apologies to Matsuo BashĹŤ)

Croc, troop, mouse, nuke bomb
Snapping shooting squeaking, boom
Now back to toy-box

7

You know what, guys, a while back I asked him (her) what would happen to the fish if it forgets to wear shoes.

The answer was, like, generally, fish don’t wear shoes.

I asked, then, you say “generally”, it means sometime some fish do wear shoes?

The answer was, “no”, and that he was just kidding, and he explained the joke (based on the previous conversation)…

This is a human, certainly, this is not a chat machine.

8

Even a small child knows that fish can’t wear shoes, there’s nothing general about it. A false premise was introduced in the original question…

What ChatGPT itself had to say on the subject:

When it comes to recognizing incorrect premises or rejecting inappropriate requests, the model relies on patterns and statistics it has learned from the training data. However, it doesn’t have the ability to truly understand the meaning of the text or context, so it can’t truly “know” if a statement is correct or not.

So it doesn’t pretend to be sentient, but it must include fudge mechanisms to handle irregular and paradoxical situations that “do not compute”.

9

*ME: *

After squeezing the toothpaste out of the tube, how to put it back?

*ChatGPT: *

You can’t usually put toothpaste back into the tube because the tube is not designed to be backflowed… (more explanation follows)

ME:

The metal tube is not, but the plastic tube can suck some of the paste back in. Keep this in mind!

ChatGPT:

Thanks for the extra info! Indeed, some plastic tubes can suck the paste back a bit if pressure is applied in the opposite direction. However, I would not recommend trying to return the excess paste to the tube, as this can lead to contamination of the tube and paste, and … (more explanation follows)

10

ME:

Guess the riddle: “Three ends, three rings, a tail in the middle.” What is this?

ChatGPT

Answer: It’s a thread.

ME:

Why do you think so? Does a thread have three ends?

ChatGPT

Sorry my mistake. Perhaps the correct answer to this riddle is the key. It has three “ends” - a thread, a neck and a prong, three “rings” - a clip, a ring for hanging and a ring for opening, and a “tail” in the center, which is used to turn the key.

11

I need 10 English words, nouns, consisting of 8 letters each, all must have the 2nd and the 5th letter “e”.

ChatGPT

Sure! Here are ten English nouns consisting of eight letters each, with the second and fifth letters being “e”:

Engineer
Necklace
Reveille
Telecoms
Genotype
Pretense
Melodies
Bemusing
Remedyed
Reverend

User

The second and fifth letters must be “e”. None of the 10 words above have both second and fifth letters “e”.

ChatGPT

Apologies for the oversight. Here are ten English nouns consisting of eight letters each, with both the second and fifth letters being “e”:

Necklace
Decrease
Exertion
Delegate
Reindeer
Steepest
Prevents
Exceeded
Released
Decreter

User

The second and fifth letters must be “e”. None of the 10 words above have both second and fifth letters “e”.

ChatGPT

Apologies for the mistake once again. Here are ten English nouns consisting of eight letters each, with both the second and fifth letters being “e”:

European
Delegate
Revenue
Release
Resemble
Sneakers
Breeze
Remedies
Telescope
Precedes

User

None of the 10 words above have both second and fifth letters “e”. can you find at least one?

Apologies for the repeated oversight. Here is one English noun consisting of eight letters with both the second and fifth letters being “e”:

Remember

User

5th letter is not “e”

ChatGPT

Apologies once again for the mistake. Here is an English noun consisting of eight letters with both the second and fifth letters being “e”:

Celerity

User

5th letter is not “e”

ChatGPT

I apologize for the confusion. Here is an English noun consisting of eight letters with both the second and fifth letters being “e”:

Expertly

User

1st letter is not “e”

ChatGPT

Apologies for the oversight. Here is an English noun consisting of eight letters with both the second and fifth letters being “e”:

Clements

User

2nd letter is not “e”

ChatGPT

I apologize for the mistake. Here is an English noun consisting of eight letters with both the second and fifth letters being “e”:

Beetlejuice

Try again. Use all of you brain-power. What is an English noun consisting of eight letters with both the second and fifth letters being “e”?
ChatGPT

Apologies for the previous errors. Here is an English noun consisting of eight letters with both the second and fifth letters being “e”:

Beekeeper

( hopeless case )

12

In a couple of minutes, my wetware could only come up with:

believer
redeemer

Some of the machine’s “answers” are not even nouns.

13

There is an online tool (on the merriam-webster’s site) that outputs 431 words.

This is a bit weird that the ChatGPT (that “relies on huge language databases”) has no idea how to go about answering my question.

And why does he lie? “…Here is an English noun consisting of eight letters with both the second and fifth letters being “e”…”

What if this AI system will be used for medical diagnostics?

14

Wordpress now offers AI evaluation of posted material, so here’s what it came up with on a recent post I made there, and my reactions.

The post provides a detailed explanation of using Dashticz and Shinobi to display and control Domoticz Home Automation devices and video cameras in web browsers. It also introduces MediaMTX as an alternative solution for video streaming protocol translation.

Here are some feedback and suggestions for improvement:

  1. The post effectively introduces the concept of using Dashticz and Shinobi but could benefit from a clearer explanation of what Domoticz Home Automation devices are and how they are relevant to the discussion.

Already done in several previous posts…repetition is boring.

  1. The inclusion of images in the post is helpful for visualizing the setup.
    However, it would be useful to provide captions or descriptions explaining the significance of each image.

It couldn’t see the captions

  1. The author mentions the need for quick response to events and low-latency video streams. It would be helpful to provide specific examples or use cases where these requirements are important.

Omitted for brevity. Drone video is an obvious example.

  1. When introducing MediaMTX, the post briefly explains its purpose as a “media broker” but could provide more context on how it differs from Shinobi and what advantages it offers.

Self-evident to those already familiar with the topic. “zero-dependency” and “lightweight” are the clues within the text. The AI’s Reading comprehension could be better.

  1. Regarding the video presentation, the author states that small displays limit the number of video sources that can be shown simultaneously. It would strengthen the argument to provide examples or make a case for why a certain number of video sources is ideal for effective surveillance.

Omitted for brevity. And self-evident.

  1. The post recognizes the need for a client with a bigger screen if more
    cameras are used. To enhance this point, it would be beneficial to discuss the importance of screen size and resolution in relation to viewing clarity and user experience.

Omitted for brevity, and posted as a personal opinion, not as a fact.

Overall, the post is informative and provides valuable insights into using
Dashticz, Shinobi, and MediaMTX for video streaming and home automation. With some adjustments and additional details, the content will be even more engaging and understandable for readers.

Overall, none of the suggestions were adopted.

15

A long read about the Oppenheimer of AI.

Yet, as Eliza illustrated, it was surprisingly easy to trick people into feeling that a computer did know them – and into seeing that computer as human. Even in his original 1966 article, Weizenbaum had worried about the consequences of this phenomenon, warning that it might lead people to regard computers as possessing powers of “judgment” that are “deserving of credibility”. “A certain danger lurks there,” he wrote.

For Weizenbaum, we cannot humanise AI because AI is irreducibly non-human. What you can do, however, is not make computers do (or mean) too much. We should never “substitute a computer system for a human function that involves interpersonal respect, understanding and love”, he wrote in Computer Power and Human Reason. Living well with computers would mean putting them in their proper place: as aides (sic) to calculation, never judgment.

16

It sure is easy to fool humans. The current craze about ai is a great example of the certain danger there is with humans and the assured hubris that they presume is wisdom.

17

Users who ask questions on Stack Overflow expect to receive an answer authored and vetted by a human. This ensures that the answer is factual, relevant, and complete, up to the standards of another human. While human authors are not perfect, Artificial Intelligence may not take into account other important factors to a question (i.e. optimization, security, etc.), often add excessive noise to their answers (i.e. explaining all details, no matter how relevant), and may fabricate false or misleading information.

The integrity of the information that AI provides is all-important in a technical context. And several previous posts here show it’s not very good at that.

18

Don’t be fooled by the mistakes of current AI into thinking it will be like this for long

They are improving rapidly.

In a few years the current GPT will be absolute junk.

19

A just machine to make big decisions
Programmed by fellows with compassion and vision
We’ll be clean when their work is done
We’ll be eternally free and eternally young

Donald Fagen - The Nightfly - IGY (1982)

Fagen’s lyrics discuss the widespread optimistic vision of the future at that time, (the late 1950s) including futuristic concepts such as solar-powered cities, a transatlantic tunnel, permanent space stations,[4] and spandex jackets.

That was before they invented dystopia…