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 30128 times)

0 Members and 1 Guest are viewing this topic.

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9890
  • Country: us
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #125 on: March 17, 2023, 05:33:12 pm »
For the big name standardized tests like the ACT and SAT, there are restrictions on calculators with CAS (Computer Algebra System) and the proctor can activate a timer on many such calculators to lock out the functionality for some period of time.

https://www.google.com/search?q=calculators+with+CAS+exam+lockout

It's complicated...

Many universities are walking away from both the ACT and SAT

https://www.wkbn.com/news/local-news/nearly-80-of-colleges-moving-away-from-act-sat-testing-for-admission/

 
The following users thanked this post: MK14

Online MK14

  • Super Contributor
  • ***
  • Posts: 4540
  • Country: gb
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #126 on: March 17, 2023, 05:41:31 pm »
True. But, unlike with calculators, chatbots have the potential to spew across all aspects of our life. (Counterpoint: what proportion of spreadsheets contain significant errors?)

The way I see it, and taking it from an electronics point of view.  Let me give you an actual (hypothetical) example.  Especially as ChatGPT (and any similar derivatives or competition), improve, their accuracy, usefulness and quality, in the future.

Let's say, someone on this forum, has just completed an electronics project they designed (hardware and/or software), and they are going to release it onto their hobby blog, hackaday, university, work environment, or to their customer.

Let's also say, they have decided (or been requested), to provide a detailed circa 20 page, instruction and technical reference manual, for it.

The options would be, the person spends (very approximately, actual time varies widely between individuals, and how busy they are with other activities) 1 week, writing it, and typing it into their favourite text or word processor.

Another option, would be to use ChatGPT as much as possible, which could reduce the time to create it down to 1 or 2 days (theoretically).

So, the author, chooses the 10 or so, chapters/sections they want, some of which could be described to ChatGPT, and allow it to write out that particular chapter or section.  Saving a lot of time and effort.

E.g. Instead of writing about not leaving batteries in it for a long time of disuse, don't mix and match different brands of batteries, and only use quality alkaline batteries.

You can say to ChatGPT:

Please write a detailed section on using 4 AA batteries, which go in the battery box, shouldn't be left in it for long periods of disuse, as they may leak.  Only use alkaline batteries, don't mix different battery types or brands, and dispose of all batteries, safely.

ChatGPT's response:

Quote
Using 4 AA batteries in a device can be a convenient and cost-effective way to power it. However, it's important to follow a few guidelines to ensure safe and effective use.

Firstly, it's important to use only alkaline batteries with the device. Alkaline batteries have a longer lifespan and are less likely to leak than other types of batteries, such as zinc-carbon batteries. In addition, different types of batteries may have different voltage and chemical properties, which could damage the device or cause it to malfunction.

Secondly, it's important not to mix different types or brands of batteries. Mixing batteries with different voltage or chemical properties can cause a variety of problems, including reduced performance, shortened lifespan, and even leakage.

Thirdly, it's important not to leave the batteries in the device for long periods of disuse. This is because over time, the batteries can leak, causing damage to the device and potentially harming the user. To prevent this from happening, it's a good idea to remove the batteries from the device when it's not in use and store them in a cool, dry place.

Finally, it's important to dispose of all batteries safely. Batteries contain chemicals that can be harmful to the environment, so it's important to dispose of them properly. Many communities have recycling programs that accept batteries, so be sure to check with your local recycling center to see if they accept batteries.

In summary, using 4 AA batteries in a device can be a convenient and cost-effective way to power it. However, it's important to use only alkaline batteries, not mix different types or brands of batteries, remove the batteries from the device when it's not in use, and dispose of all batteries safely. By following these guidelines, you can ensure safe and effective use of your device while minimizing the risk of harm to yourself and the environment.

Which then (ideally), can be tidied up by the author, and proof-read for any possible errors. Re-requesting that ChatGPT makes changes or corrections, as part of that process.

A person with limited morals, and who doesn't worry about copyright etc.  Could just google something and then just cut and paste it in, but that is another matter.
 

Online Bud

  • Super Contributor
  • ***
  • Posts: 6915
  • Country: ca
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #127 on: March 17, 2023, 05:54:23 pm »
"It is important to.." and "It is important not to..." appears to be a distinctive signature of Chat GPT.
Look how many times it used it here and in other posts.
Facebook-free life and Rigol-free shack.
 
The following users thanked this post: MK14

Online MK14

  • Super Contributor
  • ***
  • Posts: 4540
  • Country: gb
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #128 on: March 17, 2023, 06:14:07 pm »
Is that the right and best measuring stick for it?
Such lack of realization, doesn't seem to stop powerful chess playing computers, from beating, even the worlds best human players.
Or AI self driving cars, (when enabled/allowed, which sometimes happens, already, in some places, some of the time), driving me to where I need to go.

My post was about "education". But ok lets switch topics, there is no real natural "intelligent" cars driving around. Chess is not a measurement tool of human intelligence , nothing intelligent in remembering vast number of sequential pattern combinations dictated under certain rules. Go watch interviews with Garry Kasparov talking geo/politics the dude is a complete moronic fool.

Ok, taking your first point.

Firstly, sorry about that.  You're right.  Although I was originally fully following the thread.  It changed direction and there were lots of education, related posts, which I only quickly glossed over.  So, sorry again, for any confusion on my part.

I agree with you.  What you are basically saying, is that ChatGPT (and similar, modern such systems), are NOT generally intelligent, as such, that the user(s), interact with.

Which is a pity.  As I can imagine, having great fun with it, if it was generally intelligent, and could (genuinely) learn new facts and skills, as the conversations, progress.  Over a period of many hours, days, weeks  etc.

Just like we can, when dealing with the same, real life human being.

But from a psychological point of view, when I use ChatGPT.  I do sometimes get some (possibly slight or more) feelings, as if it was a real human, behind its outputs.

A bit like, feeling sympathy for a totally fictional character, in a movie, who has just come to some harm, by a bad guy, in the same movie.  Even though, it is completely make believe and fictional.
« Last Edit: March 17, 2023, 06:18:25 pm by MK14 »
 

Online MK14

  • Super Contributor
  • ***
  • Posts: 4540
  • Country: gb
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #129 on: March 17, 2023, 07:44:48 pm »
"It is important to.." and "It is important not to..." appears to be a distinctive signature of Chat GPT.
Look how many times it used it here and in other posts.

Interesting observation.  I am finding, as I use ChatGPT more and more.  It does give one a sense/ability, to kind of detect when some text, is apparently from ChatGPT.  Even if it hasn't been specifically marked as such.

Maybe its neural network, tends to see things, more as a binary black or white situation.  Rather than being able to effectively, see varying levels of grayness, allowing a smoother and better balanced answer, than strict binary 1's and 0's, true/false or yes and no, type of answers.

A bit like the old term (that seems to have somewhat gone out of fashion, these days), the so called 'fuzzy logic' system.
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 8180
  • Country: fi
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #130 on: March 17, 2023, 08:11:57 pm »
"It is important to.." and "It is important not to..." appears to be a distinctive signature of Chat GPT.

It's generating boilerplate. What I find weird that text like produced by ChatGPT was shunned just maybe a decade or two ago. This is all textbook example of poor writing; too verbose, with poor signal to noise ratio; polar opposite of "in a nutshell" writing. Yet even before the emerge of ChatGPT, I have noted such verbose writing style becoming highly popular. Clearly ChatGPT has learned that style by being trained by mostly 10-year old or newer material.
 
The following users thanked this post: MK14

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 7747
  • Country: ca
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #131 on: March 17, 2023, 09:08:28 pm »
Looking at this ChatGPT-4 review, it has greatly improved in some areas.
In fact, it now supports uploaded images.
 :scared:  @2:11 - Take a look as it can discern that an improper VGA plug was attempted to be plugged into a USB C phone port.
Has anyone here uploaded schematics and asked ChatGPT-4 to describe what we are looking at...


« Last Edit: March 17, 2023, 09:12:46 pm by BrianHG »
 

Offline EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37750
  • Country: au
    • EEVblog
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #132 on: March 17, 2023, 09:15:34 pm »
The same thing will happen with large language models.  Teachers will have to adapt how they assess their students, but not including AI in the teaching syllabus now is (IMO) a terrible mistake.

It's only been a few months since ChatGPT was released, but already practically every university and school has a policy for it now. I don't think I've ever seen a piece of technology force the word to react so quickly. Name another?
 
The following users thanked this post: tom66

Offline EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37750
  • Country: au
    • EEVblog
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #133 on: March 17, 2023, 09:16:08 pm »
A new forum section now exists, I'll move this thread over to there.
https://www.eevblog.com/forum/chatgptai/
 
The following users thanked this post: RoGeorge, MK14

Online MK14

  • Super Contributor
  • ***
  • Posts: 4540
  • Country: gb
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #134 on: March 17, 2023, 09:58:28 pm »
A new forum section now exists, I'll move this thread over to there.
https://www.eevblog.com/forum/chatgptai/

That's great!
Thanks.

I appreciate, that many people (from what I've gathered, from reading about ChatGPT), don't like, and can't stand reading or seeing, lots of output quotes.  From the ChatGPT system (and presumably similar systems).

So, by having our own area, we can keep off other peoples lawns, and away, from the (majority probably), that DON'T want to see such things, or even discussions about it.

As this (and other threads), seem to show.  Forum users, individual opinions on ChatGPT, vary widely.  From it being the next best thing since sliced bread, to it being terrible and a waste of time.

It will be interesting to see how that new child board (or sub-forum), succeeds, in practice.
 

Online Bud

  • Super Contributor
  • ***
  • Posts: 6915
  • Country: ca
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #135 on: March 17, 2023, 10:02:28 pm »
"It is important to.." and "It is important not to..." appears to be a distinctive signature of Chat GPT.

It's generating boilerplate. What I find weird that text like produced by ChatGPT was shunned just maybe a decade or two ago. This is all textbook example of poor writing; too verbose, with poor signal to noise ratio; polar opposite of "in a nutshell" writing. Yet even before the emerge of ChatGPT, I have noted such verbose writing style becoming highly popular. Clearly ChatGPT has learned that style by being trained by mostly 10-year old or newer material.
Yes but you'd not use that construction in literally every single part of a response.

Edit: I do not discount that this may be done to imply authority, i.e. "I know more than you are and therefore you should listen. "
This has a good potential to go far, doesn't it.
« Last Edit: March 17, 2023, 10:04:57 pm by Bud »
Facebook-free life and Rigol-free shack.
 

Online xrunner

  • Super Contributor
  • ***
  • Posts: 7522
  • Country: us
  • hp>Agilent>Keysight>???
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #136 on: March 17, 2023, 10:08:19 pm »
"It is important to.." and "It is important not to..." appears to be a distinctive signature of Chat GPT.

It's generating boilerplate. What I find weird that text like produced by ChatGPT was shunned just maybe a decade or two ago. This is all textbook example of poor writing; too verbose, with poor signal to noise ratio; polar opposite of "in a nutshell" writing. Yet even before the emerge of ChatGPT, I have noted such verbose writing style becoming highly popular. Clearly ChatGPT has learned that style by being trained by mostly 10-year old or newer material.
Yes but you'd not use that construction in literally every single part of a response.

It will be much harder to tell using the next version - GPT4. It can be told in what manner to respond to inquiries.

Quote
For example, GPT-4 can assume a Socratic style of conversation and respond to questions with questions. The previous iteration of the technology had a fixed tone and style. Soon ChatGPT users will have the option to change the chatbot's tone and style of responses, OpenAI said.

Read more at:
https://economictimes.indiatimes.com/news/how-to/whats-chat-gpt-4-everything-you-should-know-about-ai-that-not-only-answers-but-questions/articleshow/98679818.cms?utm_source=contentofinterest&utm_medium=text&utm_campaign=cppst
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

Online Bud

  • Super Contributor
  • ***
  • Posts: 6915
  • Country: ca
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #137 on: March 17, 2023, 10:24:48 pm »
Socratic style of conversation is also a signature.
I think a mature human will be able to pinpoint telltale signs of a machine in a conversation.
Facebook-free life and Rigol-free shack.
 

Offline Smokey

  • Super Contributor
  • ***
  • Posts: 2597
  • Country: us
  • Not An Expert
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #138 on: March 17, 2023, 10:34:28 pm »
Socratic style of conversation is also a signature.
I think a mature human will be able to pinpoint telltale signs of a machine in a conversation.

Until they train it in that style.....
 

Online xrunner

  • Super Contributor
  • ***
  • Posts: 7522
  • Country: us
  • hp>Agilent>Keysight>???
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #139 on: March 17, 2023, 10:36:13 pm »
Socratic style of conversation is also a signature.
I think a mature human will be able to pinpoint telltale signs of a machine in a conversation.

Oh that's but one example of a style.

It can already look at the EEVBlog and look at members and their posts (up to 2021). Someone has already shown this. It has even characterized individual EEVBlog members styles and interests. So I'll just tell it to post in the style of "Bud" on the EEVBlog.

Me: I would like to generate a response in the style of "Bud" on the EEVBlog, can you do that?

ChatGPT: I'd be happy to give a response in the style of Bud on the EEVBlog! What would you like me to say?

Simple as that.
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

Online MK14

  • Super Contributor
  • ***
  • Posts: 4540
  • Country: gb
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #140 on: March 17, 2023, 10:41:46 pm »
Socratic style of conversation is also a signature.
I think a mature human will be able to pinpoint telltale signs of a machine in a conversation.

Oh that's but one example of a style.

It can already look at the EEVBlog and look at members and their posts (up to 2021). Someone has already shown this. It has even characterized individual EEVBlog members styles and interests. So I'll just tell it to post in the style of "Bud" on the EEVBlog.

Me: I would like to generate a response in the style of "Bud" on the EEVBlog, can you do that?

ChatGPT: I'd be happy to give a response in the style of Bud on the EEVBlog! What would you like me to say?

Simple as that.

Too tempting for me.  Question for ChatGPT:
Quote
I would like to generate a response in the style of "Bud" on the EEVBlog forums.  Please write a long winded note, that explains how you messed up your project, by 'mistake', one rainy Saturday night, while attempting to build an interesting project.

Quote
Well, let me tell you about the time I royally messed up a project by mistake. It was a rainy Saturday night, and I was feeling adventurous. I had this idea for an interesting project that I wanted to build, and I had all the parts laid out in front of me.

I had done my research, and I felt confident that I knew what I was doing. I had read through the datasheets and schematics, and I had a clear plan in my head. I was ready to get started.

I started by carefully soldering together the components, making sure to double-check every connection. Everything seemed to be going smoothly, and I was feeling pretty proud of myself.

But then, disaster struck. As I was soldering one of the components, I accidentally touched the iron to the wrong spot. There was a loud pop, and a puff of smoke came out of the project. I knew instantly that I had made a grave mistake.

I quickly unplugged the project, but it was too late. The damage was done. I had fried a crucial component, and the project was now useless.

I was devastated. I had spent so much time and effort on this project, and now it was all for nothing. I tried to salvage it, but there was just no way. I had to admit defeat.

Looking back on it now, I realize that I made a rookie mistake. I was so focused on getting the project done quickly, that I wasn't paying close enough attention to what I was doing. I let my guard down, and I paid the price.

But, as they say, you learn from your mistakes. And I certainly learned from this one. I now take my time when working on projects, and I always double-check my work before moving on. It may take a bit longer, but it's worth it to avoid making the same mistake twice.

So, there you have it. That's the story of how I messed up my project by mistake one rainy Saturday night. It was a hard lesson to learn, but I'm grateful for the experience. And who knows, maybe one day I'll revisit that project and get it right the second time around.
 
The following users thanked this post: xrunner

Online xrunner

  • Super Contributor
  • ***
  • Posts: 7522
  • Country: us
  • hp>Agilent>Keysight>???
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #141 on: March 17, 2023, 10:47:22 pm »
Not bad! But I was really referring to GPT 4 which we can't use yet. However it will be even better than that!
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

Online tom66

  • Super Contributor
  • ***
  • Posts: 6712
  • Country: gb
  • Electronics Hobbyist & FPGA/Embedded Systems EE
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #142 on: March 17, 2023, 10:55:48 pm »
The same thing will happen with large language models.  Teachers will have to adapt how they assess their students, but not including AI in the teaching syllabus now is (IMO) a terrible mistake.

It's only been a few months since ChatGPT was released, but already practically every university and school has a policy for it now. I don't think I've ever seen a piece of technology force the word to react so quickly. Name another?

Smartwatches are probably the only other one.  Before they were commonplace you could wear a watch into an exam room, at least in the UK.  After Apple or whoever it was launched the first mainstream one nothing was allowed on your wrist. It's funny because my old Casio calculator watch did have a memo function, but given it took about half an hour to enter anything in, I think it'd be quicker to revise for the actual exam.
 

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 7747
  • Country: ca
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #143 on: March 17, 2023, 11:44:18 pm »
Not bad! But I was really referring to GPT 4 which we can't use yet. However it will be even better than that!

I think next week you can, however, it does require a paid subscription.
Checkout my video here: https://www.eevblog.com/forum/chatgptai/should-chatgpt-bots-be-banned-from-the-eevblog-forum/msg4763132/#msg4763132
 
The following users thanked this post: xrunner

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 7589
  • Country: au
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #144 on: March 18, 2023, 12:55:05 am »
The same thing will happen with large language models.  Teachers will have to adapt how they assess their students, but not including AI in the teaching syllabus now is (IMO) a terrible mistake.

It's only been a few months since ChatGPT was released, but already practically every university and school has a policy for it now. I don't think I've ever seen a piece of technology force the word to react so quickly. Name another?

Smartwatches are probably the only other one.  Before they were commonplace you could wear a watch into an exam room, at least in the UK.  After Apple or whoever it was launched the first mainstream one nothing was allowed on your wrist. It's funny because my old Casio calculator watch did have a memo function, but given it took about half an hour to enter anything in, I think it'd be quicker to revise for the actual exam.

Even the earlier "pocket" programmable calculators were pretty clunky to set up likely programs in, to the extent that after an initial "knee jerk" banning reaction, this was widely recognised, & they were allowed.

Those which had separate plug-in modules which could be "pre-programmed" were banned for longer, but over time, it was realised that they also offered little advantage.

In the main, I just used my calculator where, before, I would have used Mathematical Tables (I never mastered the Slide Rule), or just "Mk 1 inbuilt squishy memory unit".
 

Online Bud

  • Super Contributor
  • ***
  • Posts: 6915
  • Country: ca
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #145 on: March 18, 2023, 01:07:18 am »
@MK14 ha-ha! Any idiot on EEVBlog who is familiar with Bud's posts  can tell that nonsense was not written by Bud   ;D
Facebook-free life and Rigol-free shack.
 
The following users thanked this post: MK14

Offline EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37750
  • Country: au
    • EEVblog
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #146 on: March 18, 2023, 02:43:26 am »
Me: I would like to generate a response in the style of "Bud" on the EEVBlog, can you do that?
ChatGPT: I'd be happy to give a response in the style of Bud on the EEVBlog! What would you like me to say?

Much fun to be had here on the forum with this!  :-DD

Reminds me of a prolific user on Usenet electronics (and other) groups called Rod Speed.
He was so well known for writing bot-like replies that someone actually wrote a Rod Speed simulator, and it was pretty hard to tell the diffence!
 
The following users thanked this post: thm_w, MK14

Offline EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37750
  • Country: au
    • EEVblog
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #147 on: March 18, 2023, 02:45:56 am »
It will be interesting to see how that new child board (or sub-forum), succeeds, in practice.

If it's a fad and people lose interest in talking about it then no harm at all. We've mearged "dead" sections before.
 
The following users thanked this post: MK14

Online MK14

  • Super Contributor
  • ***
  • Posts: 4540
  • Country: gb
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #148 on: March 18, 2023, 03:30:09 am »
Me: I would like to generate a response in the style of "Bud" on the EEVBlog, can you do that?
ChatGPT: I'd be happy to give a response in the style of Bud on the EEVBlog! What would you like me to say?

Much fun to be had here on the forum with this!  :-DD

Reminds me of a prolific user on Usenet electronics (and other) groups called Rod Speed.
He was so well known for writing bot-like replies that someone actually wrote a Rod Speed simulator, and it was pretty hard to tell the diffence!

It will be interesting to see how that new child board (or sub-forum), succeeds, in practice.

If it's a fad and people lose interest in talking about it then no harm at all. We've mearged "dead" sections before.

I remember you saying about the notorious Rod Speed, in a past thread.  That sounds very funny.   :-DD

In the 1980's (as someone else mentioned), there was lots of hype about artificial intelligence.  But in practice, not a lot, happened.  Especially not as regards any products or software, that consumers could obtain or buy.

But this new, upcoming generation of AI stuff, with the ChatGPT, hitting the headlines, and popularity charts.  Does seem to be a very significant, new era, in computing and probably world changing event.

In my opinion, it can read random text so well, and produce rather sensible looking results and responses.  That it can be very successful.  Especially if later and later versions/generations of it, keep on improving in leaps and bounds.

It does make mistakes (too often), and has very limited real (general intelligence), at the moment.

But as time goes on, and these things get developed.  Who knows, where it may end up?

E.g. The original 'mobile phones', that were the size of shoe boxes, cost a small (or large) fortune, and may have needed extra big boxes full of electronics to allow them to work, and power them.  But, they were a genuine start of a new revolution, in mobile (phone) communications.

I suspect in coming decades (I very much hope, sooner, rather than later), we will be able to get human sized, robots, with essentially human like intelligence, who can help out and do chores and things, around peoples homes.

But on the other hand, if they can eventually, also tackle virtually any (previously paid jobs, done by humans).  That could cause major upheavals in the job market, and real big complications.  As to how wealth is earnt and distributed.

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 suspect it is trickiest, for younger people, about to go to university.  What is the best way of handling possible future changes, as regards AI.  Is there any point in going for things like programming, which may be partially taken over by automatic program generating software/AI.

There are disagreements, as to if the other skills and work, involved in real life programming work, especially within electronics embedded stuff.  Will keep such people in employment.

I think, there is a huge amount of, extra software which could be written, if cheaper options were available, to get programming work done.  So, to an extent, the market could well, still cry out for experienced programming people, such as software engineers.

I think electronic engineering, is too fiddly in some aspects, to be rapidly taken over by ChatGPT.  E.g. Who sits by a non-working, new prototype and debugs and troubleshoots it, when it doesn't work as expected.

Also some areas, are too ill defined, such as RF.

But straight forward web development work, could in theory, be partially replaced by automatic programming agents, where a normal (non-web development) worker, specifies what the business wants, then ChatGPT etc, goes ahead and implements it.
But it would still need testing and maintenance, and some aspects are best done by experienced web people, who know what they are doing.  So it's not clear, what is going to happen, in the future.

But very complicated web development, would still probably need real web developers.  E.g. Banking websites.

Maybe it will end up like wordprocessors.  Where something like 2 to 5 typists, could be replaced, by one wordprocessing operator.  Or even the manager, types the thing themselves, using a PC, and laser prints it.  But there are still many other office staff, just their precise functions are different, these days.
E.g. Instead of typewriting, or operating a wordprocessor, they might answer many emails, answer the phone, make tea/coffee, etc etc.
« Last Edit: March 18, 2023, 03:33:43 am by MK14 »
 

Offline Nominal Animal

  • Super Contributor
  • ***
  • Posts: 6266
  • Country: fi
    • My home page and email address
Re: Should ChatGPT bots be banned from the EEVblog Forum?
« Reply #149 on: March 18, 2023, 03:52:59 am »
As a typical Finn, the one thing I actually do wonder about ChatGPT, is what it thinks of me.

Not enough to create an account, though.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf