security Does NamePros allow bots to make posts?

[Compassion]

Hello members.....


With the onslaught of bot accounts on venues such as Youtube, Reddit and major news outlet sites, the question in regards to if Namepros allows bots to make posts is reasonable. as bots proliferate around the world.

Is there currently a policy allowing/now allowing bots on NP?

 

[oldtimer]

These AI Entities that live in the Cloud ... Please tell me more about this.

It's the merging of machine learning abilities of AI with the environment that the cloud computing technology provides

Just do a Google search for:

Cloud Based AI

Which in my opinion will lead to the logical evolution path of:

"AI Entities that live in the Cloud"

IMO

 

[Future Sensors]

I tried to start some discussions around this subject in the past, but there weren't too many participants.


https://www.namepros.com/threads/artificial-intelligence-domainer.1116246/


https://www.namepros.com/threads/fu...ed-and-maintained-by-ai.1151026/#post-7366359

Thank you for the references.

Personally I would not use the often overhyped terms 'AI' and 'cloud' in this context, just a 'structured bot' in a separate, controlled, Q&A/training/course like section. It may use curated, trusted data sources from namePros, some other brilliant external sources, and of course almost all previous research done by @Bob Hawkes as input

Proposing anything that goes further than this, especially in free-form debates on this forum, doesn't seem like a good idea to me. Not even in the Break Room.

 

[oldtimer]

Thank you for the references.

Personally I would not use the often overhyped terms 'AI' and 'cloud' in this context, just a 'structured bot' in a separate, controlled, Q&A/training/course like section. It may use curated, trusted data sources from namePros, some other brilliant external sources, and of course almost all previous research done by @Bob Hawkes as input

That sounds like a good idea

Proposing anything that goes further than this, especially in free-form debates on this forum, doesn't seem like a good idea to me. Not even in the Break Room.

Well when the AI threads here went inactive I kind of realised that too.

 

[Paul]

As @Bravo Mod Team has explained, there are multiple types of bots. We do not allow the types of bots you are concerned about, @Compassion: we don't allow bots to participate in discourse.

And we can publicly see this:

Right now:
The data is:

"1,226 people and 538 robots" are on NP

Those are not the types of bots you think they are. Those are mostly crawlers and RSS readers: they're used to populate search engines and news aggregators. They do not (and cannot) register or post content. If they were to try, they'd receive an error message on a blue page indicating that their traffic is considered high-risk. Some of you may have seen that error before if you tried to use a VPN that was also used by bots or otherwise tried to post on NamePros from a non-residential connection.

For example, Domaining.com has software that pulls our latest blog posts and shows links on their site. NameBio has software that pulls the latest auctions for their search engine. Both of those are bots, but they make no attempt to post content or pass as human. I think we can all agree that such bots are beneficial.

For the remainder of this post, I'm going to use the term "bot" to refer to bots that are attempting to post content and pass as human. However, this is a misleading use of the term. At NamePros, we usually use the terms "bot" and "robot" in the technical sense; I'm going to forego that for this post.

Is there currently a policy allowing/now allowing bots on NP?

We do not allow these types of bots on NamePros, and we go to great lengths to stop them.

i actually interacted with one, once.

When i realized it was a bot, i invited Mods dm.

They just “15% restricted her” Pathetic.

That wasn't a bot, although there's often little difference between a bot and a human with bad intentions. Spam humans are often more incoherent than bots.

The majority of the seemingly-automated traffic that actually manages to post on NamePros is by underpaid humans in developing nations trying to make a living by spamming. This is difficult to block because they're not bots.

We also do not allow members to go around accusing each other of being bots. If you suspect you're dealing with a bot, report the messages. We can very easily see on our end whether you're dealing with a bot.

I know bots have been utilized since I've been a member.

No, they haven't--not in the way you're thinking.

And yes interacting with bots is a reality.

For now, on NamePros, it doesn't need to be.

There are many times I have encountered accounts here that seem to behave like bots as their posts don't actually make sense in regards to the topics at hand - hence the questions.

Those are, for the most part, people. Bots based on technology like GPT-3 are better than humans at exactly two things:

1. Staying on topic
2. Sounding coherent

Humans, on the other hand, are absolutely terrible at #1, and #2 is a crapshoot.

Platforms like Twitter and Facebook are very different from NamePros in that:

• You can essentially shout into the abyss and still go viral; you don't actually need to defend your arguments. There's a lot of emphasis on the first post in any given thread, but very little attention is given to follow-up debate.
• There are also complex algorithms to determine what's trending.

That's an environment in which bots thrive. A human can author the first post and it can be spammed all over the place by bots. The bots can then engage in coherent follow-up with respondents; it doesn't matter that there's no substance to the replies, since they're usually short and most viewers will never read them. It just boosts engagement, an effective way to game ranking algorithms.

NamePros has traditional threads and tends to sort content via simple ordering (usually the date of the last reply). For now, bots can't really survive in such an environment while simultaneously making disruptive claims--and that's without taking all of our anti-bot measures into account, which make it far more difficult.

To reiterate a point I made previously: we absolutely will not tolerate members accusing each other of being bots. Report the content in question. We can very easily see on our end whether the content is automated.

If you have questions about a particular member, and you think they're a bot, feel free to DM me and I can verify. However, I can mostly guarantee that all members participating in the Break Room with any degree of regularity are bona fide humans. I know: you're skeptical. You'd love to believe other humans couldn't possibly be so ignorant, so misguided. No real human would possibly believe all of that nonsense or be so incoherent. I assure you, though, if you were to buy a plane ticket and head to their location, you'd find a human. (Please don't actually do that.)

Just in case it wasn't clear the first two times--and because I know it's going to happen no matter how many times I warn people--do not dismiss other members' content as being posted by a bot. You are interfacing with fellow humans here, and we will not tolerate people accusing each other of being anything less.

I would love to go on a rant here about how often I see two people with different world views becoming increasingly skeptical that the other could possibly be human. Yes, you're both human. Yes, you have very different views of the world. Get over it.

Pop quiz: how many times have I cautioned against publicly accusing other members of being bots?

 

[Compassion]

As @Bravo Mod Team has explained, there are multiple types of bots. We do not allow the types of bots you are concerned about, @Compassion: we don't allow bots to participate in discourse.



Those are not the types of bots you think they are. Those are mostly crawlers and RSS readers: they're used to populate search engines and news aggregators. They do not (and cannot) register or post content. If they were to try, they'd receive an error message on a blue page indicating that their traffic is considered high-risk. Some of you may have seen that error before if you tried to use a VPN that was also used by bots or otherwise tried to post on NamePros from a non-residential connection.

For example, Domaining.com has software that pulls our latest blog posts and shows links on their site. NameBio has software that pulls the latest auctions for their search engine. Both of those are bots, but they make no attempt to post content or pass as human. I think we can all agree that such bots are beneficial.

For the remainder of this post, I'm going to use the term "bot" to refer to bots that are attempting to post content and pass as human. However, this is a misleading use of the term. At NamePros, we usually use the terms "bot" and "robot" in the technical sense; I'm going to forego that for this post.



We do not allow these types of bots on NamePros, and we go to great lengths to stop them.



That wasn't a bot, although there's often little difference between a bot and a human with bad intentions. Spam humans are often more incoherent than bots.

The majority of the seemingly-automated traffic that actually manages to post on NamePros is by underpaid humans in developing nations trying to make a living by spamming. This is difficult to block because they're not bots.

We also do not allow members to go around accusing each other of being bots. If you suspect you're dealing with a bot, report the messages. We can very easily see on our end whether you're dealing with a bot.



No, they haven't--not in the way you're thinking.



For now, on NamePros, it doesn't need to be.



Those are, for the most part, people. Bots based on technology like GPT-3 are better than humans at exactly two things:

1. Staying on topic
2. Sounding coherent

Humans, on the other hand, are absolutely terrible at #1, and #2 is a crapshoot.

Platforms like Twitter and Facebook are very different from NamePros in that:

• You can essentially shout into the abyss and still go viral; you don't actually need to defend your arguments. There's a lot of emphasis on the first post in any given thread, but very little attention is given to follow-up debate.
• There are also complex algorithms to determine what's trending.

That's an environment in which bots thrive. A human can author the first post and it can be spammed all over the place by bots. The bots can then engage in coherent follow-up with respondents; it doesn't matter that there's no substance to the replies, since they're usually short and most viewers will never read them. It just boosts engagement, an effective way to game ranking algorithms.

NamePros has traditional threads and tends to sort content via simple ordering (usually the date of the last reply). For now, bots can't really survive in such an environment while simultaneously making disruptive claims--and that's without taking all of our anti-bot measures into account, which make it far more difficult.

To reiterate a point I made previously: we absolutely will not tolerate members accusing each other of being bots. Report the content in question. We can very easily see on our end whether the content is automated.

If you have questions about a particular member, and you think they're a bot, feel free to DM me and I can verify. However, I can mostly guarantee that all members participating in the Break Room with any degree of regularity are bona fide humans. I know: you're skeptical. You'd love to believe other humans couldn't possibly be so ignorant, so misguided. No real human would possibly believe all of that nonsense or be so incoherent. I assure you, though, if you were to buy a plane ticket and head to their location, you'd find a human. (Please don't actually do that.)

Just in case it wasn't clear the first two times--and because I know it's going to happen no matter how many times I warn people--do not dismiss other members' content as being posted by a bot. You are interfacing with fellow humans here, and we will not tolerate people accusing each other of being anything less.

I would love to go on a rant here about how often I see two people with different world views becoming increasingly skeptical that the other could possibly be human. Yes, you're both human. Yes, you have very different views of the world. Get over it.

Pop quiz: how many times have I cautioned against publicly accusing other members of being bots?

I appreciate the in depth response.

I'm well aware that the bots here have been and currently are of the nature of the RSS Reader and Crawler types.

I have almost no social media presence since 2019. Haven't opened FB or IG since then and my Twitter account has almost no usage, however I understand that climate has drastically changed in the recent two years and bots/AI entities are more common now while censorship is incredibly real there (even of credentialed doctors speaking facts and truths)


It looks as if accounts that post automatically/as a bot or an "AI entity that lives in the Cloud" 100% are not allowed. Period. Yes?




The reality is AI entities are currently in the process of making many human roles obsolete.
This is one reason the world economy has been intentionally stalled to a grinding halt and money velocity is slow.

Again I appreciate your transparent response.

We are all enduring a most unprecedented "World Reset" .... Let us be sure to land our intentions on compassion, truth and patterns that ripple ease and abundance.

 

[Future Sensors]

I'm gonna unsubscribe from this thread for a while, because this will become the same kind of thread we've seen elsewhere.

 

[Paul]

It looks as if accounts that post automatically/as a bot or an "AI entity that lives in the Cloud" 100% are not allowed. Period. Yes?

The oversimplified answer is "yes." As always, once we dig into the technical details, there are some exceptions.

For example, we have a pretty cool bot that posts to a hidden thread on NamePros at regular intervals. You can't see any of its posts, and it always posts the same thing, but we use it to ensure that we haven't broken anything. If it's ever unable to post, it pages whoever's on call (usually me).

Here's the actual text of its post:

Don't edit or rename this thread! {Sentinel}

We also have another bot built into the forum software that notifies moderators when it detects certain kinds of suspicious activity. Think of it as a bot to detect other bots. It doesn't post directly, but it does file reports. (Those reports are clearly marked as being from our bot.) Again, it doesn't post anything publicly--that wouldn't be appropriate.

Theoretically, there are some types of content that we might permit to be posted automatically, such as aggregated datasets posted at regular intervals. The bot would have to identify itself, and this would be handled on a case-by-case basis.

We do not allow bots to actively participate in discussions or debates, and I can't think of any scenario in which we'd make an exception to that.

 

[oldtimer]

We do not allow bots to actively participate in discussions or debates, and I can't think of any scenario in which we'd make an exception to that.

@Paul , Although you have made your views and stand very clear about this subject as to how you currently are going to be dealing with Bots (as you like to call them) , but I personally like to focus more on the areas that they can contribute in a more positive and productive way to the forum (or the domain Industry at large) and as such do you ever see a time in the near future that you might be allowing non humans to participate in the forum with all the privileges that are extended to the regular members and do you think that such ideas and possibilities should be discussed now or do you rather if we all stayed away from this subject altogether.

There are certain areas that an AI might be able to excel in like for example when it comes to giving us better appraisals and valuations for our domain names or giving us suggestions for good domains that we can hand register.

IMO

 

[Paul]

do you ever see a time in the near future that you might be allowing non humans to participate in the forum with all the privileges that are extended to the regular members and do you think that such ideas and possibilities should be discussed now or do you rather if we all stayed away from this subject altogether.

While there's a lot of hype over AI lately, until AI is close to being able to pass a Turing test, these discussions can be misleading if they give the impression that they're relevant to the present day. Part of the issue is that most people don't really understand how these technologies function or what stage they're at; that makes it difficult to have a discussion that isn't rife with misunderstandings and outright FUD.

I'll give you a quick summary of where the industry is at:

• The state-of-the-art in text-based AI--the sort that would be relevant to NamePros--is GPT-3.
• GPT-3 is adept at churning out coherent text when it isn't being scrutinized. In one instance, it managed to produce an article that was highly upvoted on a news aggregation site.
• GPT-3 doesn't really know what it's saying; given a prompt, it spits out text that seems coherent and on-topic. It's not capable of fact-checking or really "thinking" about what it's saying: it imitates language, not humans.
• As an anecdote, a friend of mine made a chat bot powered by GPT-3. While it had no shortage of snarky remarks, it wasn't really capable of conducting a conversation or debate. Its first few messages would sound coherent (albeit lacking in substance), but it would quickly start devolve into a state of confusion and repetition when asked to continue the discussion. It always remained somewhat coherent and on-topic, but it was distinctly different from talking to a human--it still feels like the chat bots of old.
• While we've come a long way from the IRC bots based on Markov chains, we're nowhere near the point of creating bots that can pass as human when challenged.

I still hold hope that we may see AI capable of passing the Turing test in my life time, but there remains a strong possibility that it won't happen.

I understand these claims may be at odds with hopeful commentary you've heard elsewhere, or with reporting in both mainstream and niche media.

For some of you, I'm sure these claims I've made go in one ear and out the other. It's very easy for you to prove me wrong: make a bot that can pass a Turing test. If you don't possess the skills to make such a bot, hire someone else to do it.

 

[redemo]

@Future Sensors 1171:4000 that's an impressive ratio!

 

[Compassion]

While there's a lot of hype over AI lately, until AI is close to being able to pass a Turing test,

I feel you are on the right track, though Wikipedia is not a reputable source and isn't permitted as a source per my graduate school studies.



I feel AI has surpassed the Turing test a lot longer ago than some may think.

I agree with this presenter that human beings are by definition more an emotional being than a logical being....

Can an AI already pass itself off as a human expert? IMHO - Yes.


One video about this topic:
(three years old)

And then there is all the hype with "sex robots" which are already live in action. Would these robots pass the Turing test?

 

[redemo]

"I feel you are on the right track, though Wikipedia is not a reputable source and isn't permitted as a source per my graduate school studies"

@Compassion I strongly agree with you. There are too many falsehoods perpetuated on Wikipedia. I have tried to correct clearly obvious lies there, only to be met with a spiteful army of Wiki rulebook nerds. It's generally a good starting place but I would never quote Wikipedia in any scientific study.

 

[Paul]

Wikipedia is not a reputable source

They're links for people to understand the terminology and are not intended as sources. If you want to learn more, you should research the topics on your own. I've linked to primary or secondary sources--not ternary sources like Wikipedia--where a source was warranted.

and isn't permitted as a source per my graduate school studies.

Hopefully Joe Scott's YouTube videos weren't permitted, either. His videos are entertaining but do not constitute an authoritative source.

I feel AI has surpassed the Turing test a lot longer ago than some may think.

I appreciate your confidence in my profession.

What's great about the Turing test is that it's quite easy to prove when you've passed it--it's an objective test.

Prove me wrong! DM me and we can arrange a double-blind trial. You supply the bot, I'll supply the willing victim (myself).

Until someone's willing to put their money with their mouth is and prove that they could run a bot on NamePros that passes the Turing test, this hypothetical discussions aren't really worth the bother.

 

[oldtimer]

I still hold hope that we may see AI capable of passing the Turing test in my life time, but there remains a strong possibility that it won't happen.

You might be right as far as the fact that we might have to wait a while longer to be able to benefit from the full potentials of AI, but we are in transformational and transitional times and things can develop much faster than we can anticipate and for that reason sometimes the lines between hype and reality might get blurred as something that sounds like hype this year might become reality in just a few years later.

Nevertheless you are sitting on a wealth of Big Data that is contained in this forum which represents the collective knowledge of the domaining community and which might come in handy if and when you decide to use it to train the NamePros AI.

I have been trying to be helpful by introducing certain new ideas to the community here, but I do respect the fact that at end NamePros is a business and as such the responsibility ultimately falls on your shoulders to decide what is in yours and NamePros' interest,

Going forward I try not to be too disruptive with this or any other subject.

IMO

 

[Lox]

@PaulThere are certain areas that an AI might be able to excel in like for example when it comes to giving us better appraisals and valuations for our domain names or giving us suggestions for good domains that we can hand register.

IMO

You mean the artificial intelligence "software" that enables f.e. namestudio.com to analyze effect of word to word frequency, infer the meaning of each word in a sentence and all the surrounding words as well? That's not AI.

In order to build something like domain name appraisal / naming software, I will need a trillions of data ... + patterns in registrants activity [f.e. search query - typing input, hot jar, etc] provided by registrars and registries > not gonna happen) and on-the-fly structured transformer-model machine learning + cover-base capable of delivering - dialogue.

Only if a whole Domain name / naming industry come together and collect f.e. (start) $200m. We can talk about it.

Today, "AI based" brand name idea generator, naming engine is just a bunch of data stuffed into whatever database and upon request delivered to the front via IF this-than-that-otherwise = this-that *priorities .ext etc etc. Any valuator / appraisal engine company using sort of machine learning ... they'll need more info than "sold for", "keyword used", "reg. in number of ext" etc.

No data, No AI.

Regards

 

[Paul]

You might be right as far as the fact that we might have to wait a while longer to be able to benefit from the full potentials of AI, but we are in transformational and transitional times and things can develop much faster than we can anticipate and for that reason sometimes the lines between hype and reality might get blurred as something that sounds like hype this year might become reality in just a few years later.

There are plenty of ways we can benefit, but don't expect a fully-interactive bot just yet.

If you have proposals for ways in which NamePros could benefit from AI, we're all ears, but the best proposals probably won't involve bots that act like humans. We already have humans for that.

Nevertheless you are sitting on a wealth of Big Data that is contained in this forum which represents the collective knowledge of the domaining community and which might come in handy if and when you decide to use it to train the NamePros AI.

I have been trying to be helpful by introducing certain new ideas to the community here, but I do respect the fact that at end NamePros is a business and as such the responsibility ultimately falls on your shoulders to decide what is in yours and NamePros' interest,

While I'm a programmer and security professional, not a sociologist or psychologist, I do worry about the current state of the relationship between our society and technology. As technology improves, some of the biggest hurdles we will encounter won't be with the technology itself, but in how we develop and codify ethics and policy surrounding it. These are tough problems to solve, and we've only seen the tip of the iceberg.

While I'm personally not opposed to using AI in a way that benefits the community, it has to be done in a responsible manner, and that's not easy to achieve if the AI is interacting directly with users. I don't have a good solution for that, but we're certainly open to suggestions.

Going forward I try not to be too disruptive with this or any other subject.

Disruptive ideas are great! Since this particular topic happens to overlap with my area of expertise, I'm attempting to ensure the discussion moves forward in a manner that facilitates debate over features an policies that are likely to be relevant in the near-term while simultaneously dispelling fear that might exist surrounding the topic.

NamePros hasn't seen truly intelligent AI yet. Even if it exists somewhere out there, most of the automated posting we see doesn't even make an attempt to be particularly deceptive. As it stands, we don't expect it to be a near-term concern, although that could always change.

This is certainly a topic you're free to discuss here, but I do have a few requests:

1. Please don't turn this into a witch hunt. If you suspect that another member is a bot, and you'd like to discuss the matter with me personally, DM me. The thread needs to remain constructive, not accusatory. (I'm not saying this has been an issue in the thread yet, but I suspect it will become an issue.)
2. We already have humans who act like humans. The best uses of AI will probably be to handle tasks that humans can't, not to impersonate humans. Please try to find uses for AI that fill gaps that can't better be filled by humans.
3. Please do your best not to promote fear. This can be a scary topic for a lot of people, especially those who aren't dealing with the limitations of computers on a daily basis as part of their job. By extension, this means the following types of statements aren't constructive:
   
   1. "Bots are running rampant and spreading misinformation on platform X, therefore they must be doing the same thing on NamePros."
   2. "I, for one, welcome our new robot overloards."
4. Avoid logical fallacies in your arguments.

 

[oldtimer]

You mean the artificial intelligence "software" that enables f.e. namestudio.com to analyze effect of word to word frequency, infer the meaning of each word in a sentence and all the surrounding words as well? That's not AI.

In order to build something like domain name appraisal / naming software, I will need a trillions of data ... + patterns in registrants activity [f.e. search query - typing input, hot jar, etc] provided by registrars and registries > not gonna happen) and on-the-fly structured transformer-model machine learning + cover-base capable of delivering - dialogue.

Only if a whole Domain name / naming industry come together and collect f.e. (start) $200m. We can talk about it.

Today, "AI based" brand name idea generator, naming engine is just a bunch of data stuffed into whatever database and upon request delivered to the front via IF this-than-that-otherwise = this-that *priorities .ext etc etc. Any valuator / appraisal engine company using sort of machine learning ... they'll need more info than "sold for", "keyword used", "reg. in number of ext" etc.

No data, No AI.

Regards

Aren't you repeating back to me what I just said myself.

 

[Lox]

Aren't you repeating to me what I just said myself.

Just clarifying so it can help other users/readers think about.

 

[oldtimer]

1. Please don't turn this into a witch hunt. If you suspect that another member is a bot, and you'd like to discuss the matter with me personally, DM me. The thread needs to remain constructive, not accusatory. (I'm not saying this has been an issue in the thread yet, but I suspect it will become an issue.)

All good points,

but

I am not sure if you are addressing the above request to me or is this for the members in the forum in general,

For the record I have never accused any other member here of being a Bot and I am sure that I am not going to start doing so now.

IMO

 

[johnn]

Would you consider my wife is bot if she posts on my behalf?

 

[Paul]

I am not sure if you are addressing the above request to me or is this for the members in the forum in general,

Just in general.

Would you consider my wife is bot if she posts ion my behalf?

I’m sorry @johnn, but we don’t allow shared accounts. I’m going to have you ban you.

 

[Compassion]

You supply the bot, I'll supply the willing victim (myself).

Would you consider my wife is bot if she posts on my behalf?

Curious minds wonder if one of the $1 million Japanese do-everything-a-wife-can robots can operate a computer and post on NP?

 

[oldtimer]

Just clarifying so it can help other users/readers think about.

I meant as far as the Big Data that can be used to train AI

non of the examples that you have mentioned are true AI

But that doesn't mean that someday soon the domain Industry can't have its own AI that has the collective knowledge and expertise of the domaining community as a base to use for reaching a higher level of knowledge and expertise of its own.

IMO

 

[johnn]

Ok Paul. I will tell her that you will ban me. Expect to receive a phone call from her.

 

[oldtimer]

Please keep your wives and companion robots out of these discussions,

We are trying to build the ultimate AI domainer here.