Should ChatGPT bots be banned from the EEVblog Forum?

[EEVblog]

I think I'm with the majority that chat bot who's sole purpose is to respond here should be banned.
Although I think it might be fun to have a new dedicated section called "Ask The Bot" where you can ask questions and get a bot reply within minutes. Basically lets us experiment and have fun with the bot within the forum.
I presume this is possible with some sort of script that could run on the server, poll for new questions, ask a pre-paid ChatGPT account, take the output and post it back into the thread?

If enough people think that could be fun then I can look into finding someone to code such a thing.

[vk6zgo]

Forum(s), could offer something on the lines of a user-ID verification process.  In effect, to confirm they are human.  Their real life (personal) details, would still be secret from other users.  But they would have been verified.

That would tie in with the UK (and other governments) desire for "digital identities". It would presumably work as well as personal digital certificates[1], and use the same mechanisms as porn sites will have to use (?credit cards?).

Or not, as the case may be

[1] I had a late colleague who, at the turn of the millennium when the government first proposed such things, pointed out to them that "My name is Donald Duck and I have a certificate to prove it".

Quite a few years back, my sister & her husband bought some property through a Real Estate Agent named Don Duck.

[MK14]

I think I'm with the majority that chat bot who's sole purpose is to respond here should be banned.
Although I think it might be fun to have a new dedicated section called "Ask The Bot" where you can ask questions and get a bot reply within minutes. Basically lets us experiment and have fun with the bot within the forum.
I presume this is possible with some sort of script that could run on the server, poll for new questions, ask a pre-paid ChatGPT account, take the output and post it back into the thread?

If enough people think that could be fun then I can look into finding someone to code such a thing.

One very quick, easy and FREE option, which would give very rapid, zero cost results.

Is to create a new forum section (or sub-forum).  E.g. A Child board, where the existing "Child Boards: Embedded Computing , Vintage Computing, Programming, Networking & Wireless , Security " ones are.

One is for discussion of ChatGPT, AI and related things.

The other has an experimental 'Automatic answers by ChatGPT' section.  But, initially it is operated manually, by users putting their questions to ChatGPT and posting the results.

Then we can see how it works out.  If it is very popular, it can be done for real, by writing script(s).

If no interest in it is shown, or it has problems.  The ChatGPT script(s) project can be abandoned and/or revised in the light of the early test results.

Anyway, a section for ChatGPT and AI things, may be a worthwhile new child board (or sub-forum), anyway.

[MK14]

Question for ChatGPT:

How to write a script for simple machines forum (SMF), which takes any threads in a special area and uses APIs to submit them to ChatGPTs API calls?

ChatGPT's response:

Writing a script to automate the process of submitting threads from a special area in SMF to ChatGPT's API can be accomplished using a few steps:

    Retrieve the threads from the special area of SMF:

You can use SMF's built-in functions to get a list of threads in the special area you want to work with. You can use the SMF API to connect to the forum database and retrieve the thread data.

    Parse the thread content:

Extract the content of each thread, including the title and body, and save them in a format that can be used by the ChatGPT API. This may require cleaning up the content, removing HTML tags, or any other necessary formatting.

    Submit the thread content to ChatGPT:

Use the ChatGPT API to submit the thread content for processing. You can use the REST API to make HTTP requests and get responses in JSON format. You may need to authenticate your API requests using an API key or other authentication method provided by ChatGPT.

    Retrieve the response from ChatGPT:

After submitting the thread content to ChatGPT, you can retrieve the response, which will be the generated text. You can parse this text and use it to update the corresponding thread in SMF.

    Update the thread in SMF:

Using the SMF API, you can update the thread with the generated text from ChatGPT. This may require creating a new post or updating the existing thread content.

Overall, the process involves retrieving thread data from SMF, submitting it to ChatGPT, and then updating the thread in SMF with the generated text. The specific implementation details will depend on the SMF and ChatGPT APIs you are using, as well as your programming language of choice.

Next question for ChatGPT:

Can you list an example program?

ChatGPT's response:

Certainly! Here is an example Python script that demonstrates the steps outlined in my previous response:

import requests
import json
import smfapi

# Connect to the SMF API and retrieve the threads from the special area
smf = smfapi.SMF("http://yourforum.com/api.php", "your_api_key")
threads = smf.get_special_area_threads()

# Loop through each thread and submit it to ChatGPT
for thread in threads:
    # Extract the thread content
    title = thread['title']
    body = thread['body']

    # Format the data for ChatGPT
    data = {
        'text': body,
        'model': 'your_model_name',
        'prompt': 'your_prompt_text',
        'temperature': 0.5
    }

    # Submit the data to ChatGPT's API
    headers = {'Authorization': 'Bearer your_api_key'}
    response = requests.post('https://api.openai.com/v1/engines/davinci-codex/completions', headers=headers, data=json.dumps(data))

    # Parse the response from ChatGPT
    generated_text = response.json()['choices'][0]['text']

    # Update the thread in SMF with the generated text
    smf.update_thread(thread['thread_id'], {'body': generated_text})

[Nominal Animal]

If Microsoft Copilot and OpenAI were forced to use proprietary code in training their language models, say the Windows and Microsoft Orifice source code, they for sure would object like hell.  But, because it is licensed under different conditions, it's okay, eh?  Fuckers.

And I thought Microsoft had changed in the last decade or so.  Nope, still love to exploit others for their private gain, just like before.

[MK14]

Everyone knows ChatGPT can be prompted to word its questions as any character with any fictional history, language skills, level of typo density, etc, right?

Just checking.

Here's ChatGPT's, response, when prompted to do that, to your post:

"Jus chkin if ChatGPT can be promptd 2 wrd its quesshuns lyk any charactr wit any fictiunal histury, langwage skillz, lvl of typo densitee, ect, amirite?

I meen, coz let's face it, dey can't spel 4 toffee, der grammer is teribl, n der punctuashun has been 4gotten."

[pcprogrammer]

if needed governments will force internet service providers to

Nowt you can do about that, but I'm buggered if I'm going to allow some random internet perve the same access.

I guess the government know all your banking details and payments. You think that should be so easily available to Dave? Or whoever hacks into Dave's user database and releases it on the Dark Web? No, the one at least has to pretend to go through the motions of getting a warrant and all that. Let Rando Webo get their own warrant if they want my personal identifying info.

I'm not saying that we should just allow every website to collect your data for verifying that you are human and increase on the risk of your information being stolen. If anything it should be less.

The whole money making business of advertising on the net is very annoying and drives me away from for instance youtube. The more adverts show up the less I will use it, and I'm not going to pay for it, because the ratio crap to usefulness is to bad.

And governments are becoming worse by the day. Take the whole cookie business. If the rule makers had actual balls they would have banned them all together, but no there is to much money involved, so lets make up some shitty rules that makes people think that we care about their privacy.

[RoGeorge]

I think I'm with the majority that chat bot who's sole purpose is to respond here should be banned.
Although I think it might be fun to have a new dedicated section called "Ask The Bot" where you can ask questions and get a bot reply within minutes. Basically lets us experiment and have fun with the bot within the forum.
I presume this is possible with some sort of script that could run on the server, poll for new questions, ask a pre-paid ChatGPT account, take the output and post it back into the thread?

If enough people think that could be fun then I can look into finding someone to code such a thing.

Sorry Dave, you are dead wrong with this, IMO.
Would be as interesting as a section with automated quotes from Google Search. 

ChatGPT is just as interesting as a 3 years old kid repeating idiom words from adults.  This gets old quick, and there is nothing more boring than posting copy paste from ChatGPT.  You are hyped because you just discovered GPT, everybody is at first.  I was too.  But it gets old fast.

Wait 1-2 more months for the hype to go away and if you still want to do that, then try it.
Halcyon, couple of replies ago,  is also very wrong in almost every sentence posted there.

The value of this forum is NOT entertainment.  It is not a Q&A like Quora, and It is certainly not a place were people are posting just to feel good either, as Halcyon thinks.  He's projecting maybe.

I think this place is nice because it's a place for alike-minded people to gather together.  Some are beginners, some are top experts of the world, but they all have in common the interest for EE, CS, science and technology in general.  They keep coming back to interact with alike minded people (not with machines).  And they come to talk about EE/CS/Sci/Tech themes.  They are not here for the circus, for the hype, or for the cooking section.

I think the poll results are very clear.
People don't want ChatGPT content.

Yet you and Halcyon think ChatGPT deserves a section in its own, and dedicated investment?!?
I don't understand how's that.  Seems illogical to me.  Maybe you see something that I don't. 

[magic]

We shouldn't be controlling what people say. If someone just wants to dribble words, I say let them. They aren't harming anyone.

Dribble sucks up some of my remaining life. While I've become good at quickly determining what to ignore, less experienced people may not have the skill/ability to do that.

We should abolish gravity. While I've become good at quickly standing up and maintaining my balance, less experience people may not have the skill/ability to do that.

[magic]

Code: [Select]

# Connect to the SMF API and retrieve the threads from the special area
smf = smfapi.SMF("http://yourforum.com/api.php", "your_api_key")
threads = smf.get_special_area_threads()

I wonder if this method actually exists

I see a problem with those AI tools so far: they rarely know what they don't know.

ChatGPT is just as interesting as a 3 years old kid repeating idiom words from adults.  This gets old quick, and there is nothing more boring than posting copy paste from ChatGPT.  You are hyped because you just discovered GPT, everybody is at first.  I was too.  But it gets old fast.

Wait 1-2 one more month for the hype to go away and if you still want to do that, than try it.

Dunno, maybe useful as a gimmick or marketing. Depends on how much it would cost to set up.

But if you are right that it's just hype, and if you wait till the hype dies down, then there will be no value in it for certain

[MK14]

Sorry Dave, you are dead wrong with this, IMO.
Would be as interesting as a section with automated quotes from Google Search. 

ChatGPT is just as interesting as a 3 years old kid repeating idiom words from adults.  This gets old quick, and there is nothing more boring than posting copy paste from ChatGPT.  You are hyped because you just discovered GPT, everybody is at first.  I was too.  But it gets old fast.

Wait 1-2 one more month for the hype to go away and if you still want to do that, than try it.
Halcyon, couple of replies ago,  is also very wrong in almost every sentence posted there.

The value of this forum is NOT entertainment.  It is not a Q&A like Quora, and It is certainly not a place were people are posting just to feel good either, as Halcyon thinks.  He's projecting maybe.

I think this place is nice because it's a place for alike-minded people to gather together.  Some are beginners, some are top experts of the world, but they all have in common the interest for EE, CS, science and technology in general.  They keep coming back to interact with alike minded people (not with machines).  And they come to talk about EE/CS/Sci/Tech themes.  They are not here for the circus, for the hype, or for the cooking section.

I think the poll results are very clear.
People don't want ChatGPT content.

Yet you and Halcyon think ChatGPT deserves a section in its own, and dedicated investment?!?
I don't understand how's that.  Seems illogical to me.  Maybe you see something that I don't. 

ChatGPT poetically says:

In this forum where we gather,
EE, CS, science and tech, all together.
A place for like-minded folks to share,
Beginners and experts, all in equal measure.

ChatGPT is an AI machine,
Providing responses, helpful and keen.
But not everyone finds it so great,
Some may prefer a different debate.

The poll results speak for themselves,
Some don't want ChatGPT on the shelves.
Yet some believe it deserves its own section,
Dedicated investment, a new direction?

It's up to the moderators to decide,
What serves the community, what they provide.
So let's keep the conversation going,
Different views, different opinions, all flowing.

Also some ChatGPT EEVblog forum jokes (reformatted by me, with A:/Q: added), to try and make amends:

Q: What do you call an EEVblog forum thread about batteries?

A: A current event!

Q: Why was the oscilloscope always causing drama on the EEVblog forums?

A:  Because it had too many triggers!

Q: Why did the EEVblog forum member cross the road?

A: To get away from the endless debates about multimeters!


Q:Why was the LED so popular on the EEVblog forums?

A: Because it always lit up the discussion!


Q: Why did the diode refuse to join the EEVblog forums?

A: Because it didn't want to be a one-way conversation!

[EEVblog]

I think the poll results are very clear.
People don't want ChatGPT content.

People don't want a bot, NOT that they don't want discussion about ChapGPT.

Yet you and Halcyon think ChatGPT deserves a section in its own, and dedicated investment?!?
I don't understand how's that.  Seems illogical to me.  Maybe you see something that I don't. 

MK14 above made a good case for a ChatGPT/AI section. Given that ChatGPT has already changed our world within a few months I think that a dedicated section is actually good idea.
It has and will impact all of us in our daily lives, our industry and our careers.
You don't have to participate, just ignore it like all the other sections you aren't interested in.

[jonovid]

if you must have one-  keep it strictly confined to electronics only topics of conversation.
do not expose it to sun light, or let it come in contact with water,  and above all, never feed it after midnight.
I've seen the movie!

[tggzzz]

I think I'm with the majority that chat bot who's sole purpose is to respond here should be banned.
Although I think it might be fun to have a new dedicated section called "Ask The Bot" where you can ask questions and get a bot reply within minutes. Basically lets us experiment and have fun with the bot within the forum.
I presume this is possible with some sort of script that could run on the server, poll for new questions, ask a pre-paid ChatGPT account, take the output and post it back into the thread?

If enough people think that could be fun then I can look into finding someone to code such a thing.

Sorry Dave, you are dead wrong with this, IMO.
Would be as interesting as a section with automated quotes from Google Search. 

<snipped supporting arguments>

Let's not miss that point; let's emphasise it.

Would be as interesting as a section with automated quotes from Google Search. 

[tggzzz]

I think the poll results are very clear.
People don't want ChatGPT content.

People don't want a bot, NOT that they don't want discussion about ChapGPT.

Yet you and Halcyon think ChatGPT deserves a section in its own, and dedicated investment?!?
I don't understand how's that.  Seems illogical to me.  Maybe you see something that I don't. 

MK14 above made a good case for a ChatGPT/AI section. Given that ChatGPT has already changed our world within a few months I think that a dedicated section is actually good idea.
It has and will impact all of us in our daily lives, our industry and our careers.
You don't have to participate, just ignore it like all the other sections you aren't interested in.

True.

But will it change our lives for the better? The jury is still out, but I'll bet it won't.

Counterexamples of how it will make our lives better...

• when a company leaks personal information, it should enable companies to quickly produce more interesting "the security of customer data is our highest priority" PR releases
• when a(nother) student is shot at university in Michigan, it will enable schools to write consoling emails better
• NewsThump often speedily skewers points, because you don't need to actually follow the URL, e.g.  https://newsthump.com/2023/03/16/government-funded-britgpt-ai-supercomputer-will-only-chat-in-three-word-slogans/

... oh. Maybe not.

[Smokey]

Prompt: What is the eevblog

ChatGPT: EEVblog (which stands for Electronics Engineering Video Blog) is a popular YouTube channel and website created by Australian electronics engineer Dave Jones. The channel focuses on electronics engineering, covering a wide range of topics including electronics design, repair, testing, and teardowns of various electronic devices.

The videos on EEVblog are known for their educational and entertaining content, often featuring Dave's unique style of presenting and humor. The channel also includes forums where electronics enthusiasts can discuss various topics and share their knowledge.

Overall, EEVblog is a valuable resource for electronics engineers, hobbyists, and anyone interested in learning more about electronics.

You can't ban ChatGPT!  It likes you, overall. 

[EEVblog]

But will it change our lives for the better? The jury is still out, but I'll bet it won't.

Entirely beside the point about whether or not to have a discussion section about it here. I think it's very worthwhile and I'll add it. if no one uses it, no big deal.

[tggzzz]

But will it change our lives for the better? The jury is still out, but I'll bet it won't.

Entirely beside the point about whether or not to have a discussion section about it here. I think it's very worthwhile and I'll add it. if no one uses it, no big deal.

Let's hope it doesn't turn into an example of "reverse broken window theory".

Here the local council has allowed some chosen street artists to put their paintings on the sides of some buildings[1]. My daughter liked the idea, I was concerned that the "other rejected" artists would vent their frustration in less legal ways. I turns out there is now a lot more unpleasant tagging around here. Shame.

[1] Even if I don't particularly like most of them, there's considerable skill displayed in many of them. But not to the degree of another local (illegal) artist, Banksy

[tszaboo]

"Now, the decision you reach here today will determine how we will regard this -- creation of our genius. It will reveal the kind of a people we are, what he is destined to be; it will reach far beyond this courtroom and this -- one android. It could significantly redefine the boundaries of personal liberty and freedom; expanding them for some -- savagely curtailing them for others. Are you prepared to condemn him and all who come after him, to servitude and slavery? Your Honor, Starfleet was founded to seek out new life; well, there it sits.; Waiting."

Ok, so to get serious, chatGPT is not AI, nor is it intelligent. It will be 1 day before 2 people writes some script to post here with it, and then before you know it, bots will spam up the whole forum talking to each other.
Do we want bots on the forum? No.
Do we want slightly more smarter bots that can maybe answer the question? Maybe? Because it's interesting to see how well it understands our issues. So I'm for the one official bot. It's also easier to shut down if it is annoying. You can program it to only reply to the original question, etc.

And BTW, I think it's just the question of time before we will not be able to tell if someone is a bot or not. You can ask GPT to write the answer like a 5 year old or a drunk.

[pcprogrammer]

A separate section to discuss AI in its entirety is completely different from having an "ask the bot" section. If there is talk about the technology behind it, it could be interesting, but just discussing ChatGPT and how it screws up, would for sure become boring real quick. And there is already at least one thread for that.

The risk I see in all this artificial intelligence is that to many people become reliant on it to think for them and society slips down even further into technology driven zombies running around. How many people don't you see roaming the streets with that "smart" phone in their hands because they can't live without them anymore.

I know this about the "smart" phone has been discussed elsewhere on the forum, but it is worrying.

I agree with RoGeorge about the reasons for being on the forum. It is for contact with like and level minded people and learn from them or discuss about technical and other  topics with them.

[tggzzz]

Ok, so to get serious, chatGPT is not AI, nor is it intelligent. It will be 1 day before 2 people writes some script to post here with it, and then before you know it, bots will spam up the whole forum talking to each other.
Do we want bots on the forum? No.
Do we want slightly more smarter bots that can maybe answer the question? Maybe? Because it's interesting to see how well it understands our issues. So I'm for the one official bot. It's also easier to shut down if it is annoying. You can program it to only reply to the original question, etc.

And BTW, I think it's just the question of time before we will not be able to tell if someone is a bot or not. You can ask GPT to write the answer like a 5 year old or a drunk.

Well, ELIZA is ~60 years old, and Parry (for paranoid, or "ELIZA with attitude") is ~50 years old. It didn't take long to get them chatting to each other.

I don't object to people giving correct answers, but I do have a problem with people giving bullshit Dunning-Kruger answers. Ditto chatbots, but I doubt the proportion of their demonstrably correct answers will be more than very low.

[PlainName]

Everyone knows ChatGPT can be prompted to word its questions as any character with any fictional history, language skills, level of typo density, etc, right?

Just checking.

Here's ChatGPT's, response, when prompted to do that, to your post:

"Jus chkin if ChatGPT can be promptd 2 wrd its quesshuns lyk any charactr wit any fictiunal histury, langwage skillz, lvl of typo densitee, ect, amirite?

I meen, coz let's face it, dey can't spel 4 toffee, der grammer is teribl, n der punctuashun has been 4gotten."

This post, and many many similar, typify the sort of post I would love to see banned from everywhere. If I wanted to know what some bot could dig up I would go and ask the bloody thing myself. I don't want every other post completely content free, merely repeating the rubbish churned out by some bot. In fact, it looks like some pre-school kid has just found something that makes them giggle and thinks adults will fund it uproarious too. No, it is just tedious attention-spamming.

Re a separate area for bot-baiting. Sure, I am all for that. But with one insta-ban rule: no crossposting to the forum proper.

[Nominal Animal]

ChatGPT reminds me of autumns in 1990s, when freshmen students all over the world got access to Unix-type machines, and discovered newsgroups, filling them with their nonsense, including "Test" posts...  I'm also old enough to have used email for a couple of blissful years completely free of mass e-mail spam.

In my opinion, it's just one more shitspout added to the world.  Moar content!  Quality does not matter!  Just make sure it looks nice and plausible!

[aargee]

What are you doing, Dave?

Daisy... Daisy... what will your answer be?

[xrunner]

What are you doing, Dave?

Daisy... Daisy... what will your answer be?

ChatGPT says:

In the movie "2001: A Space Odyssey," HAL, the advanced artificial intelligence computer aboard the spaceship, sings the song "Daisy Bell" (also known as "Bicycle Built for Two") as it is being shut down by astronaut David Bowman. This scene is a pivotal moment in the film and serves to convey the idea that HAL is not just a machine, but has developed human-like emotions and a sense of self-preservation.

The choice of the song "Daisy Bell" is significant because it was the first song ever to be sung by a computer in 1961. At the time, it was a demonstration of the computer's ability to simulate human speech. The decision to have HAL sing this song is therefore a nod to the history of computer technology and artificial intelligence.

In the context of the movie, HAL's singing of "Daisy Bell" is also symbolic of its regression to a childlike state as it is being shut down. The song is a nursery rhyme, and HAL's singing of it reflects its vulnerability and dependence on its human creators.

https://chat.openai.com/chat