Poll

Should ChatGPT bots be banned from the EEVblog Forum?

Yes, entirely. New Humans Only rule
117 (65%)
Yes, but allow one official bot that seeds replies in a special section
42 (23.3%)
No, let'em rip.
15 (8.3%)
Unsure
6 (3.3%)

Total Members Voted: 180

Author Topic: Should ChatGPT bots be banned from the EEVblog Forum?  (Read 28860 times)

0 Members and 1 Guest are viewing this topic.

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 19194
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #175 on: March 19, 2023, 09:19:03 am »
Fairly strongly in favour of banning ChatGPT, and I agree too.

But there's a problem : Stack overflow have banned it.

And nobody would want to be like stack overflow, would they ?

Miaow!
Accurate.
Shows good taste :)
Quite right.

Quote
There was a great comment :
"no wonder they banned chatgpt, because chatgpt actually tries to answer your questions instead of making fun of you, which makes them look bad."

The questions on stackoverflow are boring and only suitable when you need to do something without understanding the subtleties. Useful for "which button to I press to floggle the squirdle?", but not much more.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline artag

  • Super Contributor
  • ***
  • Posts: 1058
  • Country: gb
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #176 on: March 19, 2023, 10:09:45 am »
The questions are fine, but the moderation policy off not permitting any worthwhile discussion it what destroys it for me.
 

Offline PlainName

  • Super Contributor
  • ***
  • Posts: 6769
  • Country: va
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #177 on: March 19, 2023, 10:13:48 am »
(This one links to BigClive's Chinese lamp crimper tool video.)

Thanks! I bothered to follow that link, whereas I skipped the talking head ones, and it was a worthwhile 10 mins :)
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 19194
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #178 on: March 19, 2023, 10:58:18 am »
The questions are fine, but the moderation policy off not permitting any worthwhile discussion it what destroys it for me.

It can be difficult to separate cause from effect, and probably not profitable to try too hard.

There is one technical choice that "destroys" many forums, but not this one. Interesting subtle conversations always have a context for the to-and-fro points. Only allowing one level of context in a reply actively prevents such conversations.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline woofy

  • Frequent Contributor
  • **
  • Posts: 320
  • Country: gb
    • Woofys Place
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #179 on: March 19, 2023, 01:28:26 pm »
Well lets ask it:

Woofy:
Should ChatGPT bots be banned from the EEVblog Forum?

ChatGPT:
As an AI language model, I do not have an opinion or stance on whether ChatGPT bots should be banned from the EEVblog Forum.

However, it is important to consider the impact of such bots on the forum's comm :)unity and the rules and policies set by the forum administrators. If the use of ChatGPT bots violates any of the forum's rules or policies, then it may be appropriate for the administrators to ban them.

Additionally, it may be helpful to assess the benefits and drawbacks of using ChatGPT bots in the forum context, such as whether they enhance the user experience or hinder meaningful conversations. Ultimately, the decision to ban or allow ChatGPT bots should be based on a thoughtful consideration of these factors and the forum's community and goals.


 :)



Offline Neutrion

  • Frequent Contributor
  • **
  • Posts: 305
  • Country: hu
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #180 on: March 19, 2023, 01:37:14 pm »
I'm telling you that any benefits you think you'll see from the use of ChatGPT will be vastly outnumbered by the downsides.

What would you like me to do about it?  :-//

- Ask for AI regulation, ask to ban AI at least from:
  • autonomous AI driven weapons (in military)
  • propaganda bots (in politics)
  • over-representing and/or public human opinion-flooding by AI bots (in social media)

- Don't get the victim mindset
  • “Defeat is a state of mind; no one is ever defeated until defeat has been accepted as a reality.” (in life)

- Don't use ChatGPT when not needed
  • don't add an unnecessary ChatGPT bot/service (in EEVblog/forum)

First 3 actions that came to mind.  Pretty sure there are plenty more.



This all ChatGPT frenzy looks to me like preaching the benefits of radioactive cosmetics from the 50s.  :scared:
https://www.orau.org/health-physics-museum/collection/radioactive-quack-cures/pills-potions-and-other-miscellany/tho-radia-items.html

https://awesci.com/radioactive-toothpaste-shocking-ad-50s/

I also just wanted to suggest to have soem discussion about possible regulations. And agree with that of the victim mindset.

The whole discussin here reminds me to that one in Rivendell to decide what to do with the Ring, or what purpose it might be useful for. :)

My take:

1. Ban AI with human like interaction and communication forms.
This can only help to replace humans in their human form, and disrupt society while not solving any actual problems.

2. Ban training AI on human communication. This wil lead mostly to centralised censorship and surveillance concentrated in the hands of very few people.

3.Ban having different AI system with different sensing qualities to connect themselves. Actively prohibit it by design. (The one which can see can not talk, the one which can listen can not see et.)

Te priviledge of being able to see speak and listen and understand in a human like way should be restricted to humans. To help to protect from an overly speedy AI evolution.

For those who say banning is never a solution:
There is no industry where you can get away without any kind of regulation and safety measures. Pharmatic industry is a good example to that. So we could than also suggest the pharma industry, nuclear industry, the media, etc. to not to have any restrictions or regulations at all. This is not the way society works, not even in a small town.

As for China and the concurrence: If a strong general AI arises it will not matter where it comes from.

But until that: The uncontrolled messing around in society with AI will destroy it, I don't see now any serious counterargumnts, only wishful thinking. So the short term tech gains will be massively counterbalanced by the social collapse.

Edit:
So to make it clear these 4 points (see later for the 4.) will not ban AI completely.
We might be able to have an AI which designs for example electric circuits in the future (to be able to get rid of most of the engineers :)   But that AI will not be able to "understand" or write a joke, or be able to write code.

And the one which can write a code, would not be able to understand other human stuff.

So this is only a protection of the human domain.
« Last Edit: March 19, 2023, 03:14:23 pm by Neutrion »
 

Offline Neutrion

  • Frequent Contributor
  • **
  • Posts: 305
  • Country: hu
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #181 on: March 19, 2023, 02:12:01 pm »

There you go, you answered all your own arguments. Notice your own reasoning/wording? "seams", "able to", "like" etc. The basic course in "philosophy" gives that "real" intelligence requires
a bare minimum of self awareness, self consciousness , etc. This area of science is enormously vast.

First thing in such a philosophy course is the "reality test" (if the course is of decent standards):
1: Take your finger.
2: Look straight forward.
3: Gently press that finger tip on the side of "one" of your eye globes.

What do you see? Well you see that "reality" is in the eye of the beholder.  Now inhale/eat a magic mushroom, what is your reality now?
Not that ChatGTP ever will do any of these things "out of the blue" (another self awareness/ self consciousness thing) based on its level of human created so called artificial intelligence.

If ChatGPT is so terrible as this thread and voting portray it to be why not just let it replace all teachers on all levels of education and we will see how that will turn out.
Which is what WEF, nazi Klaus Schwab have declared they want to do. Go watch interviews whit him, read his book The Great Reset, its full of AI ChatGTP's dictating "your" life.

What about a depressed ChatGPT and Tesla self driving algorithm and banking system?



So how do you define consciousness and self awareness in general and what is tha way to test it?

And the rest: You also confuse the actual state of things with the possible outcomes in the very near future. No one suggested here that the current AI can replace completely any humans or have a btter general understanding.
Just like 5 years ago very few people thought that this level of AI would become real so fast.


There were times, long ago, when people thought that the development of machines which could churn out, thousands of properly made items, each hour.  Such as books (printing), small parts, and later bigger things.  Would cause terrible problems on the job market, as those things use to be made by humans.  But they could only produce a few items per hour, as opposed to the machines massively increased throughput.

But in practice, machines just made everyone much, much richer.  Because so many items became very plentiful and cheap.

So, a long time ago, people kept their jobs.  Just that they would tend a new fangled machine and make thousands of things, each hour.  Instead of manually making just a few items, by hand, each hour.  So, everyone was happy (especially the millionaire machine factory owners!).

So, who knows?  Maybe everything will turn out ok, in the end?


I'm not sure what all these new jobs are going to be. Maybe Douglas Adams was prescient when he foretold the shoe economy. Which neatly brings me to the origins of the word sabotage. from Wikipedia

"The English word derives from the French word saboter, meaning to "bungle, botch, wreck or sabotage"; it was originally used to refer to labour disputes, in which workers wearing wooden shoes called sabots interrupted production through different means."
https://en.wikipedia.org/wiki/Sabotage

Better make sure the Golgafrincham Ark Fleet Ship B is a big one. I have a feeling a lot of these new jobs will qualify for it. And that's actually not a good thing  regardless of the inclusion of the telephone sanitizers. You can only have so many tattoo parlors sited nearby laser tattoo removal studios and nail saloons.

I also never see any actual well defined job predictions for these new jobs.
Its only: "Well, there will be something..."
And there is a great misunderstanding in this argument with industrial revolutions before:

There the tooling the technologies and production tech.  were advancing while here the very basic human
capabilities are going to be substituted, also with non human control over them.

Perhaps the quality of the AI response will be somehow related to the quality of the request.  If that is true, highly skilled people will be required to interface with the AI.  If you want replies headed in a particular direction (eg FreeRTOS versus simple interrupt driven), you may need to nudge the AI in a direction.  But first you need to know that there is a direction and the benefits of heading in that direction.  Random walk isn't a good strategy.

"Garbage In, Garbage Out" will be a prevalent theme, IMO.


As far as studying up on this technology, it's all I've been into for a while now. Whether you like it or not, the world is going to change because of this technology, and there is no turning back. So you really have no choice but to learn, understand, and master it or get left behind.

You will not be able to outcompete this thing, as all these modells also will change really fast. (And they are orienting themself of the very quality of yours with which you would try to understand them with possibly even phisically more neurons in the future) When you think you learned something, it can be thing of the past in the next version or in another neural network.
It is black box even for it's designers. It is basically a chain reaction in the computer industry.
So I am actually really happy that Mores prediction is at its limits at least on the HW side.

And it wil be really hard to bring 99,99 procent of the population to even understand the current level of AI. So if the smartest people think that they can get away without the rest of the society than... well than they are maybe not that smart at all :)

And as for the there is no turning back: That was Sarumans argument as well :)


Oh so a fourth of extra extra restriction in the posts above:
Phisically limit the used neurons/transistors, as to not to let it get to a human level on the HW side.



« Last Edit: March 19, 2023, 02:18:40 pm by Neutrion »
 

Offline Neutrion

  • Frequent Contributor
  • **
  • Posts: 305
  • Country: hu
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #182 on: March 19, 2023, 03:21:31 pm »
Dave: This topic is so much hidden now, that I think it actively prevents a lot of....humans to chime in.
So there might be people who are not generally AI freaks to actively look for an AI section but still would read throught it, which I think is a good thing. I certainly would not have actively look for it in the AI section.
(As actually the whole issue is not funny thing to think about.)
 

Offline MK14

  • Super Contributor
  • ***
  • Posts: 4527
  • Country: gb
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #183 on: March 19, 2023, 03:42:27 pm »
Oh so a fourth of extra extra restriction in the posts above:
Phisically limit the used neurons/transistors, as to not to let it get to a human level on the HW side.

The history books, tend to show, that such big restrictions, don't necessarily work in practice.
 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9885
  • Country: us
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #184 on: March 19, 2023, 08:27:18 pm »
[I also just wanted to suggest to have some discussion about possible regulations. And agree with that of the victim mindset.


State actors (countries) will blow past any treaties or agreements if doing so will provide even a modest advantage/benefit.  They just classify the projects and screen the workers.

Individual companies will still pursue prohibited technologies at their 'skunk works' factories.  Universities are still going to disclose everything they know about technology (perhaps with licensing involved) and continue to push the limits.  What else are math majors going to do?

The cat is out of the bag and ripping through the house.  The time to prevent this kind of thing was back around 1950 when computers first became a thing.

There is so much tutorial information on the Internet that almost anybody can get involved with Machine Learning.  It isn't really necessary to understand the nits of the library code.  Just cut and paste applications and you are in the AI group.  Tweak a few hyperparameters and you're off to the races.

It's an exciting time!
 
The following users thanked this post: MK14

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9885
  • Country: us
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #185 on: March 19, 2023, 08:45:04 pm »
In the attached project, the Arduino is simply an output of facial recognition done on a PC with a webcam.  Nevertheless, facial recognition is now 'pedestrian', everybody can do it.

https://www.instructables.com/Face-Recognition-and-Identification-Arduino-Face-I/

This type of facial recognition has been used for PC login for several years so it's nothing new.  The exciting bit is that it is now at the hobby level.

 
The following users thanked this post: MK14

Offline John B

  • Frequent Contributor
  • **
  • Posts: 793
  • Country: au
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #186 on: March 20, 2023, 04:59:43 am »
If you think your ex can hold a grudge, just wait until a malicious AI has a perfect and permanent facial and identity profile database on anyone who has ever said anything negative about AI bots.  >:D
 
The following users thanked this post: MK14

Online coppercone2

  • Super Contributor
  • ***
  • Posts: 9166
  • Country: us
  • $
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #187 on: March 20, 2023, 05:35:06 am »
 
The following users thanked this post: MK14

Offline pcprogrammer

  • Super Contributor
  • ***
  • Posts: 3572
  • Country: nl
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #188 on: March 20, 2023, 05:55:17 am »
If you think your ex can hold a grudge, just wait until a malicious AI has a perfect and permanent facial and identity profile database on anyone who has ever said anything negative about AI bots.  >:D

That will come as soon as they learn it to have emotions and feelings  :palm:

It will be hateful once a month.  :-DD

Online xrunner

  • Super Contributor
  • ***
  • Posts: 7485
  • Country: us
  • hp>Agilent>Keysight>???
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #189 on: March 20, 2023, 11:02:41 am »
The AI can be your protector too!


« Last Edit: March 20, 2023, 11:04:58 am by xrunner »
I told my friends I could teach them to be funny, but they all just laughed at me.
 
The following users thanked this post: MK14

Offline Neutrion

  • Frequent Contributor
  • **
  • Posts: 305
  • Country: hu
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #190 on: March 20, 2023, 01:05:26 pm »
[I also just wanted to suggest to have some discussion about possible regulations. And agree with that of the victim mindset.


State actors (countries) will blow past any treaties or agreements if doing so will provide even a modest advantage/benefit.  They just classify the projects and screen the workers.

Individual companies will still pursue prohibited technologies at their 'skunk works' factories.  Universities are still going to disclose everything they know about technology (perhaps with licensing involved) and continue to push the limits.  What else are math majors going to do?

The cat is out of the bag and ripping through the house.  The time to prevent this kind of thing was back around 1950 when computers first became a thing.

There is so much tutorial information on the Internet that almost anybody can get involved with Machine Learning.  It isn't really necessary to understand the nits of the library code.  Just cut and paste applications and you are in the AI group.  Tweak a few hyperparameters and you're off to the races.

It's an exciting time!
I did not propose to forbid machine learning.

But you mix up two different thing: The fact that there are ways to go around regulation does not mean that regulations don't work at all, and we don't have them in any industry.
If it would work like this there would be no things forbidden, and no regulation for anything.
EVERY industry is full of regulations. Except the AI industry at the moment.
Did you check what kind of tests for how long time must be done until a medicine can be sold? It takes 15-20 years, and for good reasons.

And I think when the realisation will reach more and more citizens, there will be some regulation, but driven by the industry itself. And not on a democratic base, because those people who have some technical understanding instead leave the stuff for the very much apperciated politicians and industry lobbysts. And it will be made by politicians like the mentioned PM.

If this is the way how democratic societies are working, well, than they will be no democracies for a long time. (Just saw a report about a british small company  which will do AI filtering for social media- So it heads clearly in direction AI deciding soon what people are allowed to say without much human oversight, because humans will be expensive.)
For some people again: NO WE ARE NOT THERE YET. But heading there without much resistance. And there are no debates about this in the society.



So no regulation never worked ever for any technology, and even during the mentioned industrial revolutions were followed by a massive regulatory wave when people realized what goes wrong.
We won't have that much time for this revolution.

 

Offline Junaid_raza

  • Contributor
  • !
  • Posts: 12
  • Country: us
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #191 on: March 20, 2023, 10:02:33 pm »
We cannot deny that ChatGPT will replace humans since we will continue to rely on it and ask for assistance with everything, including personal issues.
 

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 14230
  • Country: fr
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #192 on: March 20, 2023, 10:23:33 pm »
If it replaces humans, whoever is it going to give assistance to? ::)
 

Offline EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #193 on: March 21, 2023, 01:17:29 am »
And I think when the realisation will reach more and more citizens, there will be some regulation, but driven by the industry itself.

It is absolutely guaranteed that AI will be either self or government regulated on various topics.
AI will never EVER to be allowed to tell the truth on certain topics, like politics, finance, religion, and various social issues. Much of society being functional relies on lies, and the absense of truth, or the illusion of truth.
 

Offline John B

  • Frequent Contributor
  • **
  • Posts: 793
  • Country: au
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #194 on: March 21, 2023, 02:04:34 am »
If it replaces humans, whoever is it going to give assistance to? ::)

It will be a therapist for the Bing AI, that thing is/was crazy. (It may have undergone shock therapy in the meantime)
 

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 14230
  • Country: fr
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #195 on: March 21, 2023, 04:01:17 am »
And I think when the realisation will reach more and more citizens, there will be some regulation, but driven by the industry itself.

It is absolutely guaranteed that AI will be either self or government regulated on various topics.
AI will never EVER to be allowed to tell the truth on certain topics, like politics, finance, religion, and various social issues. Much of society being functional relies on lies, and the absense of truth, or the illusion of truth.

Yep.
The question of that making society "functional" is debatable.
But lies certainly have the well known potential of avoiding/limiting conflict, and hard conflicts could get us back to wild animals.

The #1 reason we use lies is to avoid conflict. Absolutely no doubt about that.
Other living species use "lies" to the same end as well.

 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 19194
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #196 on: March 21, 2023, 09:35:38 am »
AI will never EVER to be allowed to tell the truth on certain topics, ...

"... the truth..." ?! That sounds like something that should be written in green ink.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline Neutrion

  • Frequent Contributor
  • **
  • Posts: 305
  • Country: hu
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #197 on: March 21, 2023, 02:55:28 pm »
And I think when the realisation will reach more and more citizens, there will be some regulation, but driven by the industry itself.

It is absolutely guaranteed that AI will be either self or government regulated on various topics.
AI will never EVER to be allowed to tell the truth on certain topics, like politics, finance, religion, and various social issues. Much of society being functional relies on lies, and the absense of truth, or the illusion of truth.

Yep.
The question of that making society "functional" is debatable.
But lies certainly have the well known potential of avoiding/limiting conflict, and hard conflicts could get us back to wild animals.

The #1 reason we use lies is to avoid conflict. Absolutely no doubt about that.
Other living species use "lies" to the same end as well.

I don't think that this applies generally.  I actually think, the more lies you have in a socity the less it will work, or the less people will be happy in it. As it is in private life as well.
In totalitarian systems, every communication is actually ment to be for the greater good of people, but the opposite effect is reached.

So I think the regulation must not be what it can or can not say, as it will be impossible to implement properly even with the best intention. Yes I belive all censorship in the world was created with some good intention.
So therefore it should not interfere in general human communication at all, that also to avoid a strong general AI.
 

Offline RoGeorge

  • Super Contributor
  • ***
  • Posts: 6136
  • Country: ro
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #198 on: March 22, 2023, 02:40:24 am »
If you think your ex can hold a grudge, just wait until a malicious AI has a perfect and permanent facial and identity profile database on anyone who has ever said anything negative about AI bots.  >:D

That will come as soon as they learn it to have emotions and feelings  :palm:

It will be hateful once a month.  :-DD

Emotions

 :)

(full song https://youtu.be/YnnyBFz7iUw)
 
The following users thanked this post: pcprogrammer

Offline woofy

  • Frequent Contributor
  • **
  • Posts: 320
  • Country: gb
    • Woofys Place
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #199 on: March 22, 2023, 09:48:17 am »
The most worrying thing I've found about ChatGPT is that its will confidently answer any question even when that answer is close to pure fiction.
Yesterday I asked it to provide an XOR routine for the ESP32 ULP FSM. The results were absolute rubbish. It apologized profusely when I pointed out its errors but then either rephrased the error or generated new ones. It defied all attempts to guide it, clearly having no understanding of the ULP instruction set.

It is the confidence that it produces these responses that worries me.


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf