Should all content posted anywhere on this forum as factual information require attribution?

I think the answer is yes.

We now live in a world flooded with AI generated content. Just because information comes from ChatGPT, Claude, Gemini, or any other AI model does not automatically make it accurate.

These systems can and do consistently produce hallucinations, incorrect facts, outdated information, and completely invented details presented with great confidence.

For that reason, I think any factual claims posted from an AI generated summary should also include the original source material that the AI used, assuming sources are available. And if the AI cannot produce a source for the information, then it should not be posted, or it should be clearly labelled as unverified.

Without attribution, there is really no way for anyone to verify the information independently. Readers are effectively being asked and expected to trust the AI rather than the underlying evidence.

In my view, AI generated content should not be treated as a source in itself. It should be treated as a tool that helps locate, organize, and summarize information from actual sources and then the sources need to be scrutinized.

In my own use of AI apps, I have created personal settings on all of them that force the AI models to label each piece of information they provide to me as either “Certain”, “Likely”, or “Guessing”. It makes it much easier to distinguish their outputs between fact and fiction and gives me a better idea of how much confidence I should place in what I am reading.

Otherwise, what value does the information really have?

12 minutes ago, BilllyGOAT said:

Should all content posted anywhere on this forum as factual information require attribution?

I think the answer is yes.

We now live in a world flooded with AI generated content. Just because information comes from ChatGPT, Claude, Gemini, or any other AI model does not automatically make it accurate.

These systems can and do consistently produce hallucinations, incorrect facts, outdated information, and completely invented details presented with great confidence.

For that reason, I think any factual claims posted from an AI generated summary should also include the original source material that the AI used, assuming sources are available. And if the AI cannot produce a source for the information, then it should not be posted, or it should be clearly labelled as unverified.

Without attribution, there is really no way for anyone to verify the information independently. Readers are effectively being asked and expected to trust the AI rather than the underlying evidence.

In my view, AI generated content should not be treated as a source in itself. It should be treated as a tool that helps locate, organize, and summarize information from actual sources and then the sources need to be scrutinized.

In my own use of AI apps, I have created personal settings on all of them that force the AI models to label each piece of information they provide to me as either “Certain”, “Likely”, or “Guessing”. It makes it much easier to distinguish their outputs between fact and fiction and gives me a better idea of how much confidence I should place in what I am reading.

Otherwise, what value does the information really have?

Only salvation: Every content using AI should be marked at such. (Some "mediatheks" are doing this already).

Everybody not adhering to this principle: 10 years in prison.

If not, the world will drown in uncontrolled "misinformation".

23 minutes ago, BilllyGOAT said:

Should all content posted anywhere on this forum as factual information require attribution?

I think the answer is yes.

We now live in a world flooded with AI generated content. Just because information comes from ChatGPT, Claude, Gemini, or any other AI model does not automatically make it accurate.

These systems can and do consistently produce hallucinations, incorrect facts, outdated information, and completely invented details presented with great confidence.

For that reason, I think any factual claims posted from an AI generated summary should also include the original source material that the AI used, assuming sources are available. And if the AI cannot produce a source for the information, then it should not be posted, or it should be clearly labelled as unverified.

Without attribution, there is really no way for anyone to verify the information independently. Readers are effectively being asked and expected to trust the AI rather than the underlying evidence.

In my view, AI generated content should not be treated as a source in itself. It should be treated as a tool that helps locate, organize, and summarize information from actual sources and then the sources need to be scrutinized.

In my own use of AI apps, I have created personal settings on all of them that force the AI models to label each piece of information they provide to me as either “Certain”, “Likely”, or “Guessing”. It makes it much easier to distinguish their outputs between fact and fiction and gives me a better idea of how much confidence I should place in what I am reading.

Otherwise, what value does the information really have?

Don't use AI. Use your own, if you have any.

Everything produced or assisted by AI should be clearly labeled as such.

Trying to attribute/reference everything would take an age and make the AI posts even longer.

YES....!!!

Attribution should ALWAYS be required by us true ACADEMICS....

To quote NOAM.....

EVERYBODY knows this.

34 minutes ago, BilllyGOAT said:

Should all content posted anywhere on this forum as factual information require attribution?

I think the answer is yes.

We now live in a world flooded with AI generated content. Just because information comes from ChatGPT, Claude, Gemini, or any other AI model does not automatically make it accurate.

These systems can and do consistently produce hallucinations, incorrect facts, outdated information, and completely invented details presented with great confidence.

For that reason, I think any factual claims posted from an AI generated summary should also include the original source material that the AI used, assuming sources are available. And if the AI cannot produce a source for the information, then it should not be posted, or it should be clearly labelled as unverified.

Without attribution, there is really no way for anyone to verify the information independently. Readers are effectively being asked and expected to trust the AI rather than the underlying evidence.

In my view, AI generated content should not be treated as a source in itself. It should be treated as a tool that helps locate, organize, and summarize information from actual sources and then the sources need to be scrutinized.

In my own use of AI apps, I have created personal settings on all of them that force the AI models to label each piece of information they provide to me as either “Certain”, “Likely”, or “Guessing”. It makes it much easier to distinguish their outputs between fact and fiction and gives me a better idea of how much confidence I should place in what I am reading.

Otherwise, what value does the information really have?

Yes, people need to provide sources. ChatGPT does provide sources, but I disregard information from Reddit or similar sources.

Should all content posted anywhere on this forum as factual information require attribution?

Yes, here is some examples:

In the Welcome to the World News Forum:

• Any alleged factual claims must be supported by a valid link to a mainstream media source.

In the Welcome to the Political Forum :

• Links to outside content should always include relevant commentary. Blind links to outside content with no context or explanation may be removed.

In the Welcome to "What the Papers are Saying" forum:

If you are posting a news article for discussion, please check the following:

• Its within "fair use"; that is - the title and a paragraph about the story.

• Its not already posted in World News, Soap Box or other News areas on the forum; duplication will be removed.

• Please always provide a link to your original source

In the War in Ukraine forum - welcome :

• Any alleged factual claims must be supported by a valid link to an approved, credible source.

In the The War in Israel forum:

• Any alleged factual claims must be supported by a valid link to an approved credible source.

The main reason I posted this topic is because there are now a number of users who consistently post content generated by AI as if it is providing some useful information that other people cannot just as easily find by themselves.

Personally I find much of it a waste of time and space. Most of this information is readily available and does not need to be regurgitated by another forum member.

I definitely do think AI content has its place on the forum. For example when someone is asking a specific question and another user uses AI to help find or clearly summarize information that is genuinely difficult to present otherwise. In those cases it can be very useful.

What I am seeing however is heavy overuse. It is being used either as the basis for starting new topics, or as filler by some people who cannot think of an original response on their own but still wants to have a voice in the discussion.

In most cases I see at least some effort to make it transparent via screenshots of the AI output so it is clear where it came from and readers can more easily judge it for what it is, or in my case, simply completely skip over the post.

The problem is when people copy and paste only the text from AI responses directly into posts with no reference at all, presenting it as if it is their own informed contribution. That, in my view, is the part that should be fully halted.

Even the screenshot approach becomes questionable when it is used purely as filler rather than as a direct answer to a question where someone is actually looking for help and it definitely should not be the basis of initiating a brand new topic for discussion or debate.

Otherwise it starts to feel like noise. Content for the sake of content, rather than meaningful contribution to a discussion.

No attribution should not always be required because

a lot of sources are not allowed on AN.

16 minutes ago, FlorC said:

No attribution should not always be required because

a lot of sources are not allowed on AN.

All they have to do is say that our source is not approved. It gives them an easy way to censor us for going against the party line.

49 minutes ago, BangkokHank said:

All they have to do is say that our source is not approved. It gives them an easy way to censor us for going against the party line.

Shocking! Ask for a refund mate 😄

Sure, and let's start with those movies on Amazon TV that are mostly Ai generated and look more like a video game then a movie.

21 minutes ago, SingAPorn said:

Sure, and let's start with those movies on Amazon TV that are mostly Ai generated and look more like a video game then a movie.

Our future, Never mind Steven Spielberg.

13 hours ago, BilllyGOAT said:

We now live in a world flooded with AI generated content. Just because information comes from ChatGPT, Claude, Gemini, or any other AI model does not automatically make it accurate.

These systems can and do consistently produce hallucinations, incorrect facts, outdated information, and completely invented details presented with great confidence.

I use AI because it's quick. I don't expect it will be 100% accurate, but neither is any other source. All reporters have biases, for one thing, so they put their own spin on everything.

It's mass delusion to believe your "reputable news sources" are spewing out nothing but verified facts.

You need to use your own intuition to determine what seems reasonable and what doesn't.

12 hours ago, BilllyGOAT said:

The problem is when people copy and paste only the text from AI responses directly into posts with no reference at all, presenting it as if it is their own informed contribution.

I use my Snipping Tool to copy and paste when I quote AI to differentiate it from my own text. Agree people should not blend their own words with ChatGPT and clearly identify AI text.

But AI ain't going anywhere and it's a valid source of information. On most easily verifiable non-controversial and not too complex issues, AI is useful.

13 hours ago, BilllyGOAT said:

Just because information comes from ChatGPT, Claude, Gemini, or any other AI model does not automatically make it accurate.

Somebody has been waking up to reality. Thanks for sharing something we all know!

14 hours ago, BilllyGOAT said:

I have created personal settings on all of them that force the AI models to label each piece of information they provide to me as either “Certain”, “Likely”, or “Guessing”.

Good idea. I also added:

Be direct and avoid sycophantic responses when I correct you. Acknowledge errors plainly without excessive flattery;

41 minutes ago, save the frogs said:

It's mass delusion to believe your "reputable news sources" are spewing out nothing but verified facts.

But for sure my sources are better than yours.

1 hour ago, save the frogs said:

You need to use your own intuition to determine what seems reasonable and what doesn't.

Precisely, which is why having a link to the source is essential if I'm to do my own due diligence.

15 hours ago, BilllyGOAT said:

Should all content posted anywhere on this forum as factual information require attribution?

I think the answer is yes.

We now live in a world flooded with AI generated content. Just because information comes from ChatGPT, Claude, Gemini, or any other AI model does not automatically make it accurate.

These systems can and do consistently produce hallucinations, incorrect facts, outdated information, and completely invented details presented with great confidence.

For that reason, I think any factual claims posted from an AI generated summary should also include the original source material that the AI used, assuming sources are available. And if the AI cannot produce a source for the information, then it should not be posted, or it should be clearly labelled as unverified.

Without attribution, there is really no way for anyone to verify the information independently. Readers are effectively being asked and expected to trust the AI rather than the underlying evidence.

In my view, AI generated content should not be treated as a source in itself. It should be treated as a tool that helps locate, organize, and summarize information from actual sources and then the sources need to be scrutinized.

In my own use of AI apps, I have created personal settings on all of them that force the AI models to label each piece of information they provide to me as either “Certain”, “Likely”, or “Guessing”. It makes it much easier to distinguish their outputs between fact and fiction and gives me a better idea of how much confidence I should place in what I am reading.

Otherwise, what value does the information really have?

So much for easy-going chit-chat in The Pub. Now it's 'lawyer-up or keep your mouth shut'. All a bit pretentious for such a casual setting, in my opinion.

16 hours ago, BilllyGOAT said:

Should all content posted anywhere on this forum as factual information require attribution?

I think the answer is yes.

It depends.
If the statements are presented as statements of fact - then "Yes."
If the statements are presented as statements of opinion - then "No."

3 hours ago, save the frogs said:

I use my Snipping Tool to copy and paste when I quote AI to differentiate it from my own text. Agree people should not blend their own words with ChatGPT and clearly identify AI text.

But AI ain't going anywhere and it's a valid source of information. On most easily verifiable non-controversial and not too complex issues, AI is useful.

I agree, I occasionally report some AI information and always say that is where it is from. One can then check on AI and the information I am talking about is usually cited by AI from a news article but I do not link most things I write about as it is usually from my mind and People should know how to search for information using the computer. IMHO anyway but I seldom do cite AI for information. I will try to avoid using it unless it does have some source usually.

Required? Absolutely not!

I believe that most of us use AN as a relaxation read, local color and advice, and 'some news.' When we require accuracy in reports, we can head to our favorite providers and look it up. 30 seconds of effort. Everyone now has access to the same Internet information and can easily check out the information that's important to them. No need to have it spoon-fed.

To force people to spend time and energy, half the time forgetting, just to add an attribute to something that may or may not really be accurate anyway, will stop a lot of people from posting.

6 hours ago, gargamon said:

Good idea. I also added:

Be direct and avoid sycophantic responses when I correct you. Acknowledge errors plainly without excessive flattery;

Yes, that's useful too.

I've come up with a list of 7 items I cover in personalization settings. It all seems to work best with Claude. The others get too controlled by the instructions and it makes some of their responses too unnatural. As a result I'm doing a lot more with Claude now:

1. Writing style: Avoid dashes as punctuation.

2. Role: Act as an advisor, not an assistant.

3. Response opening: Lead with challenges, blind spots, questions, or hard truths instead of agreement.

4. Confidence labeling: Mark statements by certainty level.

5. Banned phrases: Avoid specific approval and validation phrases.

6. Disagreement format: Explain why, suggest an alternative, and identify the risk.

7. Pushback handling: Stay with your position unless new evidence is provided.

6 hours ago, save the frogs said:

I use my Snipping Tool to copy and paste when I quote AI to differentiate it from my own text. Agree people should not blend their own words with ChatGPT and clearly identify AI text.

But AI ain't going anywhere and it's a valid source of information. On most easily verifiable non-controversial and not too complex issues, AI is useful.

We are at a point where some people simply abuse it all. They constantly post AI slop as if it is their own thoughts by copying and pasting words from an AI summary into a post on a topic, often multiple times on the same topic, and they add nothing else to their responses. No attribution and none of their own original opinions or content. Just low quality responses that nobody needs or wants. These individuals do it incessantly. You probably know who I'm mainly referring to. It should be stopped.

16 minutes ago, BilllyGOAT said:

We are at a point where some people simply abuse it all. They constantly post AI slop as if it is their own thoughts by copying and pasting words from an AI summary into a post on a topic, often multiple times on the same topic, and they add nothing else to their responses. No attribution and none of their own original opinions or content. Just low quality responses that nobody needs or wants. These individuals do it incessantly. You probably know who I'm mainly referring to. It should be stopped.

Reminds me of the Wizard of Oz... If I only had a brain...

4 hours ago, connda said:

It depends.
If the statements are presented as statements of fact - then "Yes."
If the statements are presented as statements of opinion - then "No."

20 hours ago, BilllyGOAT said:

Should all content posted anywhere on this forum as factual information require attribution?

The question the OP asked regarded posting as factual information.

22 hours ago, BilllyGOAT said:

We now live in a world flooded with AI generated content. Just because information comes from ChatGPT, Claude, Gemini, or any other AI model does not automatically make it accurate.

These systems can and do consistently produce hallucinations, incorrect facts, outdated information, and completely invented details presented with great confidence.

Generally I agree. What I do (regardless of this forum) when using AI on Google, is I do indeed check the sources it claims to have arrived at its summary. More often that not, those sources do not include the language of the AI summary. So in other words it hasn't been validated by the sources it includes.

5 hours ago, BilllyGOAT said:

We are at a point where some people simply abuse it all. They constantly post AI slop as if it is their own thoughts by copying and pasting words from an AI summary into a post on a topic, often multiple times on the same topic, and they add nothing else to their responses. No attribution and none of their own original opinions or content. Just low quality responses that nobody needs or wants. These individuals do it incessantly. You probably know who I'm mainly referring to. It should be stopped.

Throw some names out there. I don't see it that often but maybe I'm being fooled.

I think what you mean is to include the text of the question you asked, for example, Google's AI. I'm installing some tools so AI doesn't should up in search unless you want it to.