The artificial intelligence MEGATHREAD

[Dr. Shades]

Shades/mods - I'm going to flag this comment to bring it to your attention, and ask you to validate the following suggestion, or modify it, or tell me that it's completely wrong. If it's completely off kilter, I apologise for bothering you with it.

Here's my suggestion: Look at the Purdue University OWL - Online Writing Lab - section on Avoiding Plagiarism.

Use A.I. all you want for research, but treat AI-generated content as if it were the published words of a human author that you had read in a book. With the lessons on "Avoiding Plagiarism" in mind, in your own words, write your own paraphrase/precis of what you read, properly attributed. The extra piece of information needed here is that you must also include the name of the A.I. you used and the full prompt text.

This new text is yours, and can be used on this board outside of the A.I. megathread.

NO.

Although everyone else would be able to work with that, MG 2.0 has proven himself unable. Therefore, Malkie, for the love of God and all of His holy angels, DO NOT LEAD MG 2.0 INTO TEMPTATION.

To wit, From this point forward, DO NOT COME UP WITH WAYS FOR MG 2.0 TO USE A.I. LEGITIMATELY.

Let this be your eternal rule of thumb: As far as MG 2.0 is concerned, A.I. does not exist and never did exist. He has never heard of it in his life, nor will he ever. Not from you, not from anyone.

[Marcus]

We knew that was coming.

[MG 2.0]

Shades/mods - I'm going to flag this comment to bring it to your attention, and ask you to validate the following suggestion, or modify it, or tell me that it's completely wrong. If it's completely off kilter, I apologise for bothering you with it.

Here's my suggestion: Look at the Purdue University OWL - Online Writing Lab - section on Avoiding Plagiarism.

Use A.I. all you want for research, but treat AI-generated content as if it were the published words of a human author that you had read in a book. With the lessons on "Avoiding Plagiarism" in mind, in your own words, write your own paraphrase/precis of what you read, properly attributed. The extra piece of information needed here is that you must also include the name of the A.I. you used and the full prompt text.

This new text is yours, and can be used on this board outside of the A.I. megathread.

NO.

Although everyone else would be able to work with that, MG 2.0 has proven himself unable. Therefore, Malkie, for the love of God and all of His holy angels, DO NOT LEAD MG 2.0 INTO TEMPTATION.

To wit, From this point forward, DO NOT COME UP WITH WAYS FOR MG 2.0 TO USE A.I. LEGITIMATELY.

Let this be your eternal rule of thumb: As far as MG 2.0 is concerned, A.I. does not exist and never did exist. He has never heard of it in his life, nor will he ever. Not from you, not from anyone.

This has been an interesting back and forth on the pros and cons of A.I. usage on this board. Or should I say, the cons. One or two posters have now suggested that there could be more or less a compromise. That would have been fine by me.

I think that the overall flow of information and thought will now be restrictions through the funnel of negativism and limited points of view. One poster, as I've said recently, cannot deal with the onslaught of opinion/information that may or may not be accurate.

This is unfortunate and I think the 'power voices' have rules the day although obviously I will never be able to prove that. So be it, so let it be done. Another hot for free speech and information.

I will not add anything more than I have to this ongoing dilemma/discussion. At this point it is a fruitless endeavor...if it already wasn't from the beginning. I expect to get ragged on some more by those 'powers that be'.

So be it.

Regards,
MM

[malkie]

MG, I have to accept - and I'm sorry to say that it seems you have to accept it too - that it appears that you still don't get it, as witnessed by your statement: "One or two posters have now suggested that there could be more or less a compromise. That would have been fine by me.". I've done my best, just as I did with the issue of the single-sentence paras also. But clearly my best was not good enough.

I hope that this will also be my final word on the use of AI here as a tool for discussion.

[malkie]

Shades/mods - I'm going to flag this comment to bring it to your attention, and ask you to validate the following suggestion, or modify it, or tell me that it's completely wrong. If it's completely off kilter, I apologise for bothering you with it.

Here's my suggestion: Look at the Purdue University OWL - Online Writing Lab - section on Avoiding Plagiarism.

Use A.I. all you want for research, but treat AI-generated content as if it were the published words of a human author that you had read in a book. With the lessons on "Avoiding Plagiarism" in mind, in your own words, write your own paraphrase/precis of what you read, properly attributed. The extra piece of information needed here is that you must also include the name of the A.I. you used and the full prompt text.

This new text is yours, and can be used on this board outside of the A.I. megathread.

NO.

Although everyone else would be able to work with that, MG 2.0 has proven himself unable. Therefore, Malkie, for the love of God and all of His holy angels, DO NOT LEAD MG 2.0 INTO TEMPTATION.

To wit, From this point forward, DO NOT COME UP WITH WAYS FOR MG 2.0 TO USE A.I. LEGITIMATELY.

Let this be your eternal rule of thumb: As far as MG 2.0 is concerned, A.I. does not exist and never did exist. He has never heard of it in his life, nor will he ever. Not from you, not from anyone.

Understood. I did the best that I could, and see that, in the end, it was doomed to failure.

Thanks for your patience - you and the other mods.

[Marcus]

NO.

Although everyone else would be able to work with that, MG 2.0 has proven himself unable. Therefore, Malkie, for the love of God and all of His holy angels, DO NOT LEAD MG 2.0 INTO TEMPTATION.

To wit, From this point forward, DO NOT COME UP WITH WAYS FOR MG 2.0 TO USE A.I. LEGITIMATELY.

Let this be your eternal rule of thumb: As far as MG 2.0 is concerned, A.I. does not exist and never did exist. He has never heard of it in his life, nor will he ever. Not from you, not from anyone.

Understood. I did the best that I could, and see that, in the end, it was doomed to failure.

Thanks for your patience - you and the other mods.

Your effort is appreciated, I'm sure, but Shades' point is unassailable:

...Although everyone else would be able to work with that, MG 2.0 has proven himself unable...

[MG 2.0]

...Although everyone else would be able to work with that, MG 2.0 has proven himself unable...

I have disagreed and pushed back, but am willing to abide by final rulings. I only pushed back, well, for all the reasons I mentioned...to no avail.

Regards,
MG

[Physics Guy]

There isn't necessarily a problem, at least not that I can see, with using one of the current chatbots as a possibly more efficient Google search, to collect information and ideas. They don't always work even for this, because they still just poll the internet for patterns, and do not even try to understand anything or know what it means. I recently asked a quite specific question involving some technical terms, and the AI summary feature simply knee-jerked to some of the terms, spitting out the stuff that is most often said about them without the slightest sign of grasping my actual question. Nonetheless it can often happen, I suppose, that an AI answer to a well-posed question can turn what would have been a half-hour Google crawl into a three-second summary skim.

Some people don't just use AI as a quick source of information that they can absorb, and that they will then understand just the same as they would have if they had learned it all from thick books or direct experience. Instead it sometimes seems that the thing that people want from the AI is not facts or ideas at all, but rather a few paragraphs of slick prose that will make whatever assumptions they gave the AI sound as good as possible. That's something that the current batch of so-called AIs do even better than they find information—better, that is, in the one sense that they reliably do make things sound good. The problem is that it's also the worst thing that the current AIs do. They always make things sound good, but what they say is often absolute garbage, and they are shameless about it.

To use a current AI as a source of nice phrasing, rather than as a source of information to learn for oneself, is to think like an AI oneself, so that only the familiar patterns matter and we don't even ask about meaning or truth.

[MG 2.0]

There isn't necessarily a problem, at least not that I can see, with using one of the current chatbots as a possibly more efficient Google search, to collect information and ideas. They don't always work even for this, because they still just poll the internet for patterns, and do not even try to understand anything or know what it means. I recently asked a quite specific question involving some technical terms, and the A.I. summary feature simply knee-jerked to some of the terms, spitting out the stuff that is most often said about them without the slightest sign of grasping my actual question. Nonetheless it can often happen, I suppose, that an A.I. answer to a well-posed question can turn what would have been a half-hour Google crawl into a three-second summary skim.

Some people don't just use A.I. as a quick source of information that they can absorb, and that they will then understand just the same as they would have if they had learned it all from thick books or direct experience. Instead it sometimes seems that the thing that people want from the A.I. is not facts or ideas at all, but rather a few paragraphs of slick prose that will make whatever assumptions they gave the A.I. sound as good as possible. That's something that the current batch of so-called AIs do even better than they find information—better, that is, in the one sense that they reliably do make things sound good. The problem is that it's also the worst thing that the current AIs do. They always make things sound good, but what they say is often absolute garbage, and they are shameless about it.

To use a current A.I. as a source of nice phrasing, rather than as a source of information to learn for oneself, is to think like an A.I. oneself, so that only the familiar patterns matter and we don't even ask about meaning or truth.

Useful contribution. Thanks.

Unfortunately we will not benefit at all from any additional content and analysis that A.I. might provide. With Perplexity providing links for its sources it also provides additional exploration into what the A.I. provided. As it is, we are left pretty much to the contortions/exclusion of information that we often see posted by critics with an agenda.

Regards,
MG

[Marcus]

...Instead it sometimes seems that the thing that people want from the AI is not facts or ideas at all, but rather a few paragraphs of slick prose that will make whatever assumptions they gave the AI sound as good as possible. That's something that the current batch of so-called AIs do even better than they find information—better, that is, in the one sense that they reliably do make things sound good. The problem is that it's also the worst thing that the current AIs do. They always make things sound good, but what they say is often absolute garbage, and they are shameless about it.

To use a current AI as a source of nice phrasing, rather than as a source of information to learn for oneself, is to think like an AI oneself, so that only the familiar patterns matter and we don't even ask about meaning or truth...

Couldn't agree more. Perfectly said.

(And this: "...only the familiar patterns matter...[not] meaning or truth..." in my opinion explains why Dr. Shades is so adamant about who his new rule was made for.)