I found an AI

[LoyceV]

I'm not going to keep arguing here, I know this forum! we all love drama.

Wrong. I don't like drama and I hate chatbot shitposters. I've reported your post, and left neutral feedback.

or did you miss the "Warning - while you were reading 1 new replies have been posted. You may wish to review your post."?  

I turned that off years ago, it's annoying. But it does mean I sometimes overlook a new post.

I don't think it would be difficult to get around AI detectors, without making your post looking nonsenable.

His post just doesn't make sense at all. I'll quote the full post before it gets deleted:

Personally, I guess you are probably one of the fees that still thinks that bitcoin will suffer for its reputation having been mentioned as ransom, of course this is not the first event with bitcoin as ransom requested, however as one who is quite optimistic about public opinions, I therefore, disagree that it will have any kind of effect later on, bitcoin from the time of my knowledge has never had love from the people especially the government, so they always find a way to make it seem like the problem meanwhile the fact that the American government could protect it's citizen is the failure rather a currency that is more money than the US dollar so criminals understands what most people refuse to understand that bitcoin is quite the fastest means of cross boarder payment and with some level of knowledge they could escape track but even if they requested dollar would dollar be seen as evil as public opinion will see bitcoin of course no. This only shows that it's a figure pointed at bitcoin not today but long ago.

Even though I'm a pacifist, yet some level of crime seems to prove that the US current government is more like a racism and hate catalyst that a government with the peace in heart for the people having failed to protect it's people yet fight them using ICE and then push blames towards others for being responsible for the injustice. what a terrible   government. I hope they able to take responsibility of the security of it's citizen than fight bitcoin for no reasons.  

I especially like the "from the time of my knowledge" part.

there are valid explanations for why they were able to read that line of text.

The first thing I did was click "quote", because in my browser it looks like this:


Contrary to popular opinion, I'm not a bot, I'm human. We have no sunshine today but the sky behind the clouds is probably still blue

[bitbollo]

It's your nice catch, and you can report AI posts in another thread too
AI Spam Report Reference Thread

The member who created that thread has replied in the thread! And according to Nutildah, the content might not be AI generated. Here's a short quote of Nutildah's reply:

The reason why the post doesn't show up as AI in the detectors is because its not AI, is my guess.

To be honest, it's hard to find any connection of AI text with the posts of the Ivystar5, but surely some people who're good at using LLMs might know tricks that could bypass AI detectors.

It's hard to blame an user but this could be seen somewhat at edge I don't like post writing by AI. There is totally lack of sense, a wall of text, words that you will never use apart from ideas that not often appears to be so smart.

I would not be really surprised if some users take initial text of AI and then edit some words/phrases to reword and keep a new for as text made by humans.
In any case, it's kind of impressive to see the user has reported this term "pacifist". Anyway, in the long terms these tricks could not work anymore

[joker_josue]

I understand the idea and objective of the experience described by PrimeNumber7.

If this example was used without AI to respond, I don't know and I didn't investigate.

But the experiment is somewhat flawed from the start. If you set the text size to 1, you're not hiding it. The human eye can see that something is written there. Just use a quote and you can quickly read what's there. Something I usually do when I see very short texts (and it seems the author of this post did it too).

Out of curiosity, I think if I had replied to your topic, I would have tried to follow that "hidden rule"—I always appreciate a good sense of humor. 

Either way, I understood the idea and the objective of the experiment.

[TypoTonic]

I that was a good one brother! but well I read the thread I felt like no one is even following the op so I decided to do what he said in the thread just to be mentioned here and seeing that op was decieving people and I was the person who got decieved.

Well, if I were to follow the instruction as a joke, I'll probably add a note to avoid any misunderstanding.

If you set the text size to 1, you're not hiding it. The human eye can see that something is written there. Just use a quote and you can quickly read what's there. Something I usually do when I see very short texts (and it seems the author of this post did it too).

OP could've used [color=white] to hide it better, or maybe even [color=transparent] if he wants to fully hide it. It will still be included if someone copies the post and feeds it to an AI chatbot.
For example, something like this would be very hard to notice at first glance, unless you quote the post.

[SatoPrincess]

OP could've used [color=white] to hide it better, or maybe even [color=transparent] if he wants to fully hide it. It will still be included if someone copies the post and feeds it to an AI chatbot.
For example, something like this would be very hard to notice at first glance, unless you quote the post.

Yeah I got that idea after xLays used it on this thread lol. The text would still be visible if you quote the post or copy it into a notepad before replying. My question is, how often do people quote long posts and read it from their browser before replying?

[SamReomo]

In any case, it's kind of impressive to see the user has reported this term "pacifist". Anyway, in the long terms these tricks could not work anymore

I fear in future there might be no need of using such tricks and it would be a simple task for AI LLMs to make any generated text as fully human written text. Even, this day, using some tricks wisely can bypass all available AI content detectors.

The ones who're against use of AI content will avoid using any such tricks but it would be useful to do some tricks to show how easy it can be to bypass most AI content detectors only for demonstration purpose.

[eternalgloom]

OP could've used [color=white] to hide it better, or maybe even [color=transparent] if he wants to fully hide it. It will still be included if someone copies the post and feeds it to an AI chatbot.
For example, something like this would be very hard to notice at first glance, unless you quote the post.

Yeah I got that idea after xLays used it on this thread lol. The text would still be visible if you quote the post or copy it into a notepad before replying. My question is, how often do people quote long posts and read it from their browser before replying?

It seems those AI shit poster doesn't bother to read anything, they just copy paste other people's post and command their AI to generate new post. They really don't have the slightest respect for their Sig campaign job.

I don't think it would be difficult to get around AI detectors, without making your post looking nonsenable.

His post just doesn't make sense at all. I'll quote the full post before it gets deleted

Uh--huh, and finally he got banned, right?

Hopefully that will cut down on some of the AI junk on the forum, lol...

[vapourminer]

Out of curiosity, I think if I had replied to your topic, I would have tried to follow that "hidden rule"—I always appreciate a good sense of humor. 

Either way, I understood the idea and the objective of the experiment.

me too. if i had replied i would of put the word and the two emojies at the very beginning/end of the post or even just in a row somewhere by themselves in the middle on a newline.

like, i would do it here but i cant remember the word lol

[LoyceV]

Uh--huh, and finally he got banned, right?

If it is, it's temporary. But the 1 year signature ban looks good. I don't expect too many posts for a year without earning on that account.

[Ultegra134]

I understand the idea and objective of the experience described by PrimeNumber7.

If this example was used without AI to respond, I don't know and I didn't investigate.

But the experiment is somewhat flawed from the start. If you set the text size to 1, you're not hiding it. The human eye can see that something is written there. Just use a quote and you can quickly read what's there. Something I usually do when I see very short texts (and it seems the author of this post did it too).

Out of curiosity, I think if I had replied to your topic, I would have tried to follow that "hidden rule"—I always appreciate a good sense of humor. 

Either way, I understood the idea and the objective of the experiment.

Lol, imagine being flagged by the whole forum for messing around with the "hidden rule" and having to prove a negative. The experiment is somewhat flawed but it's going to catch users who blindly copy/paste whole topics/posts into AI generators. Someone who even bothered to even read what it had been written would likely notice it, but from our experience, how often is that by such users? Perhaps as others suggested, there might be other ways to mask it a little better, either by setting the color of the text as white or transparent.

[Ambatman]

Wow I responded to the thread I didn't even notice anything off
Had to go back and realised they were write ups that felt like lines.
I guess I'm getting old.

Perhaps as others suggested, there might be other ways to mask it a little better, either by setting the color of the text as white or transparent.

It was stated already that quoting the post brings everything out without it being hidden.

I don't think it would be difficult to get around AI detectors, without making your post looking nonsenable.

Why not use the time in making the post nonsensical to create a genuine post
Because I don't think it can get worse than what they posted
Which is why I would agree with Nutilda except humaniser can make a post that bad.

[fillippone]

Edit: You can actually see a tiny "line" on the post, but initially, I thought it was the [hr.] line or was on purpose to separe the text.

I think you could change the colour to match the background, making the instruction not visible to the human eye (unless quoting).
Anyway, I like this trick, I could start adding it  to everything I post!

[Ultegra134]

It was stated already that quoting the post brings everything out without it being hidden.

I mentioned that already in the first page of this thread in my initial post. Never claimed that it can't be seen.

I think you could change the colour to match the background, making the instruction not visible to the human eye (unless quoting).

I highly doubt that the majority of people using AI are going to bother reading what's included in the quote. Certainly, you can't catch them all, but in this case, if it wasn't for the OP, I don't think this specific user would have been spotted easily, or if he was, I'm not quite sure how possible would be to acquire substantial evidence from AI detectors that his posts were AI generated.

[SamReomo]

If it is, it's temporary. But the 1 year signature ban looks good. I don't expect too many posts for a year without earning on that account.

Here, I would add that those who misuse their technical knowledge for such tasks often end up like the that member. Although, such tactic might be hard to find at first glance but it's karma or the law of cause and effect that takes place and that's what happened with the member who used the tricks to hide that he was using AI content to get paid.

Such a misuse of AI, If I recall correctly then a high ranked member from Stake campaign also used AI content to get paid and such activity destroyed that account. The greed is the main enemy in such cases, the member's who're in campaigns where more posts get paid, such members either give more time to forum and make more posts on daily basis but that's mostly spam, or they try to trick others using AI content generation tools.

Same is true for the members running multiple accounts, but in the end they always end up making mistake of forgetting the law of cause and effect. Even by using most advanced tactics, they'll always end up getting caught and destroy their hard work. Even extra smart people are human and they sometimes make human error and that's enough to find how they broke the rules to get extra advantage. I hope the ones who've found a way to use AI such tactically will learn from the mistakes of the members who got a signature ban, and stop using such malpractices for good.

[lovesmayfamilis]

Same is true for the members running multiple accounts, but in the end they always end up making mistake of forgetting the law of cause and effect. Even by using most advanced tactics, they'll always end up getting caught and destroy their hard work. Even extra smart people are human and they sometimes make human error and that's enough to find how they broke the rules to get extra advantage. I hope the ones who've found a way to use AI such tactically will learn from the mistakes of the members who got a signature ban, and stop using such malpractices for good.

That's right, people will always make mistakes. But it seems to me that this technique, which was shown by OP, will already be taken note of by AI lovers, and the citation of the post will be checked every time. Just like all the subsequent experiments that you asked to know about as a demonstration, they will be taken into account by those who abuse AI. Didn't the same thing happen when we showed connections by wallets and other coincidences? The cheaters have become more attentive, and maybe even more arrogant, because they know the ways in which they could be tracked. In the end, he who has eyes will see, but he who has reason will cease.

[LoyceV]

Out of curiosity, I think if I had replied to your topic, I would have tried to follow that "hidden rule"—I always appreciate a good sense of humor.  

Lol, imagine being flagged by the whole forum for messing around with the "hidden rule" and having to prove a negative.

I'm pretty sure joker_josue, a human pacifist, would have done it in such a way that it's obvious he's turning it into a joke Unlike the spambot.
Just like nobody tagged me for saying blue sunshine.

Someone who even bothered to even read what it had been written would likely notice it,

Even the spammers can't get themselves to read the incoherent shitposts they create. I wouldn't even be surprised if someone has fully automated posting chatbot outputs. Each account that doesn't get banned for it is a win for them.

[joker_josue]

Edit: You can actually see a tiny "line" on the post, but initially, I thought it was the [hr.] line or was on purpose to separe the text.

I think you could change the colour to match the background, making the instruction not visible to the human eye (unless quoting).
Anyway, I like this trick, I could start adding it  to everything I post!

In the forum layout, there are two backgrounds for posts. Therefore, unless you were careful to always pay attention to the background that will be used in your post, it wouldn't always work.

I think these tricks will only create noise in the forum, as well as extra work for those who post correctly. In the short term, it would have little impact.

[BlackHatCoiner]

So basically the state of affairs in Bitcointalk right now is to write a bunch of nonsense to catch users who rely on AI bots.

Yeah, it's good initiative to detect and humiliate them, and I saw the new AI guidelines thread, but let's just be honest. What's the point? As long as posts are paid to be made, and AI technology allows for infinite text generation, there's an asymmetric incentive to just use AI in this place.

[SatoPrincess]

Same is true for the members running multiple accounts, but in the end they always end up making mistake of forgetting the law of cause and effect. Even by using most advanced tactics, they'll always end up getting caught and destroy their hard work. Even extra smart people are human and they sometimes make human error and that's enough to find how they broke the rules to get extra advantage. I hope the ones who've found a way to use AI such tactically will learn from the mistakes of the members who got a signature ban, and stop using such malpractices for good.

That's right, people will always make mistakes. But it seems to me that this technique, which was shown by OP, will already be taken note of by AI lovers, and the citation of the post will be checked every time. Just like all the subsequent experiments that you asked to know about as a demonstration, they will be taken into account by those who abuse AI. Didn't the same thing happen when we showed connections by wallets and other coincidences? The cheaters have become more attentive, and maybe even more arrogant, because they know the ways in which they could be tracked. In the end, he who has eyes will see, but he who has reason will cease.

Unfortunately, in most cases, we can’t prove the truth without providing evidence. Cheaters aren’t going to stop,  I have seen threads made by new accounts asking how to catch alt accounts lol. As we have seen in the past, these cheaters may learn and try to hide their tracks but eventually they slip up and one tiny mistake is all we need to nail them.

[joker_josue]

Yeah, it's good initiative to detect and humiliate them, and I saw the new AI guidelines thread, but let's just be honest. What's the point? As long as posts are paid to be made, and AI technology allows for infinite text generation, there's an asymmetric incentive to just use AI in this place.

I don't even know how the forum survived 10 years without AI...   

In this regard, it is up to campaign managers to ensure that they only accept users who are clearly participating in the forum, without resorting to generating AI-generated data in a generalized way.