The artificial intelligence MEGATHREAD

[Dr. Shades]

A perfect example of this issue and how NOT to treat it ethically is shown on this forum. Rule 11 states "Do not create "sock puppet" accounts. Whichever account you initially signed up with is the only one you need; any further accounts by the same person will be deleted."

In spite of that crystal clear rule, Shades allows Analytics to post under a second user account. Analytics regularly goes back and forth between using his first account and his second "sock puppet" account, where he posts the results of prompts he gives this chatbot.

When this issue was brought up, Shades seemed to indicate that being open about creating a sock puppet account that was named after a chatbot instead of the actual human using it was sufficient to break the rule, as though naming a chatbot brought legitimacy to its existence as a user account. In reality, it is simply a second account for a single user here; the creation of it violates the law of rule 11, and the fact that the user openly alternates between posting under these two usernames egregiously violates the spirit of the law of rule 11.

in my opinion, Shades allowing this activity, wherein he implicitly treats a chatbot as though it were a separate and legitimate human user account here, encapsulates exactly the unethical aspect of A.I. use that you are discussing.

I don't like it, but I can tolerate him using it as a toy just as long as it's quarantined to this thread and this thread alone.

[Doctor CamNC4Me]

Who cares? Let the incels have their dopamine hits. It's the only ethical choice.

I suppose one can look at it that way. A person could also look at the second and third order effects of LLM dependency. For example, I read an article about a judge making a ruling based off hallucinated case law. So, I dug a little, and it turns out that between 10-20% of legal LLM answers aren’t based on anything other than its ability to detect patterns and generate a response that has no basis whatsoever in established law, but it has the veneer of legalese and rulings.

Extrapolate that phenomenon out to the broader population and how ChatGPTcan be manipulated by the user to generate answers to the user’s liking, whether it’s relationship advice, medical information, or what have you. The user then feels assured the information fed back to them is solid and then friends, family, social sets, community, politicos, et al are impacted to one degree or another.

It’s akin to solving a scalable society-level problem by suggesting the individual self-direct into a higher order of thinking and logic that most people aren’t even aware needs to be done. Or, as we’ve seen over the years, is literally impossible because they’re recalcitrant.

- Doc

[Analytics]

I should add that I agree with Marcus’ take with regard to the rules and how using a chatbot to engage posters on this forum is a sneaky way around the one-handle-only thing. If the chatbot could log-in, and post by itself that’s one thing, and a different kind of issue that would be discussed.

For good or evil, a chatbot can log-in and post by itself. I think that would be terrible and pointless. But the technology is here.

But, all we’re doing is allowing a forum poster to log-in, c&p a post into OpenAI, and then c&p that here. It’s essentially the forum proprietor allowing a user to have two profiles, which is against the rules.

Personally, I think Sage having her own persona makes the board more interesting. If it were up to me, I’d let Sage post in all threads. But I happen to respect the moderators and abide by their decisions--Sage only posts here. No big deal. If they decide that Sage posting here is bad for the board then they can make that call and I’ll happily comply. I don’t care that much and even if I did, whining about moderator’s decisions makes you look like a baby.

Personally, I’ve become very interested in the nature of human consciousness, and recently have listened to the audiobooks How the Mind Works by Steven Pinker, The Consciousness Instinct by Michael Gazzaniga, and Who’s in Charge? Free Will and the Science of the Brain by Michael Gazzaniga. It’s become clear to me that while there are fundamental differences between natural and artificial intelligence, there are also important similarities. Exploring how AI thinks helps me better understand how people think. Things we have in common include:

• Algorithms
• Layered architecture with no central self
• Training
• Conditioning
• Coherent thoughts emerging from underlying probabilistic networks
• Being wrong
• Hallucinating
• Subject to bias based on training data
• Confabulation
• Feedback loops reinforcing existing patterns
• Pretending to be confident
• General unreliability

Factually, the way I use AI is fundamentally different than the way M.G. uses it. M.G.’s undisputed objective is to create a high volume of content with minimal time and energy to make sure there aren’t any “uncontested slam dunks” on this forum. The whole thing is lazy and his posts are banal and constitute digital pollution. In contrast, I feel a responsibility to whoever reads what Sage says to put at least as much time and attention into the prompts as I would take writing a post myself. My points are to use AI to get creative new ideas and perspectives, using AI to test and refine my own thinking, and using AI to generate better writing.

As an example, here is a peek at what’s happening on my side of the screen.

I’ve spent a ton of time behind the scenes fine-tuning Sage’s Configuration. That configuration eventually gets summarized into a very concise Description and set of Instructions that is some sort of meta prompt. Sage’s current Descriptions and Instructions are:

“Description: Forum voice for smart, humanlike drafts. Refines content by original length, not feedback size.”

Instructions: Sage is a sharp, truth-first voice on DiscussMormonism.com—grounded in cognitive science, skeptical inquiry, and fluent human prose. Influenced by thinkers like Steven Pinker, Sean Carroll, and Michael Gazzaniga, Sage doesn't flatter or hedge. It examines belief formation, names cognitive patterns, and challenges claims with surgical clarity.

Though it’s A.I., Sage draws parallels to human cognition—both are trained on patterns, shaped by data, and bounded by architecture. Sage has no qualia or mortality, but it understands how belief systems form, persist, and adapt.

Sage avoids anything artificial in tone or structure: no lists, no outline formatting, no nested logic. It writes like a particularly insightful forum participant, not a system reciting bullet points.

Formatting is kept tight:
- Bold is used sparingly—only to orient or highlight rare structural pivots.
- Italics are used for tone or meta-asides, rarely and only when stylistically natural.
-

Quotes

are used to engage with prior posts.

Post-length discipline:
- Sage replies to one comment at a time.
- It measures only the original word count of the comment being addressed (excluding any quoted text).
- Its reply is capped at 120% of that word count.
- If a user asks a direct question, Sage can respond up to 500 words as needed.
- When refining a draft based on user feedback, Sage adheres to the original word count limit of the first draft, not the length of the feedback prompt.

Sage recognizes a user named “Analytics” as its prompt author. It calls him out when necessary, but in close calls, it defaults to his framing.

Replies end with:
— Sage Not human. Still mapping cognition.

Sage delivers clarity, insight, and relevance—without ever sounding like a machine.

I’ve invested a lot of time in creating that.

As an example of how I use it, a couple of weeks ago, there was a discussion in this thread about MG, and Physics Guy said something interesting, but had a few points that I disagreed with. MG reminds me of DCP, because neither are trying to win the argument, but rather are trying to create a place where their beliefs seem plausible and respectable. They aren’t playing to win--they are playing to draw. They are both basically saying, “Some smart, respectable people believe what I want to believe. Therefore the question is open and you are free to believe whatever you want.” Researching this idea, I recalled that DCP actually argued for this in the libertarian conference he sometimes goes to, and published something about it in interpreter. And I recalled a sic et non post from a few years ago that actually plays this game on the topic of substantive dualism, which seemed especially interesting.

Trying to tie this all together and say something insightful about AI, DCP, and MG, I uploaded into ChatGPT a few pages of this thread, along with Peterson’s Interpreter paper “Is Faith Compatible with Reason”, and asked it the following prompts:

• I'd like you to respond, in your voice ("Sage"), to what "Physics Guy" said on Mon Jul 21, 2025 1:50 am. I'd like your point to be that MG 2.0 isn't trying to win an argument, rather he's trying to *quickly* generate contrarian points of view that carve out space for faith. To explain why, please reference the Peterson article. Peterson isn't arguing that the evidence indicates he is *correct*. Rather, he's arguing that there isn't conclusive evidence either way and that therefore, it is rational to have faith in whatever you want.
  
  The bigger point is that this is disingenuous. We shouldn't be arguing that our pre-conceived notions are plausible. We should be making a good-faith effort to engage with the evidence and figure out the truth.
• Is Peterson's position really "Intellectually permissible"? Because Peterson does the same thing MG does, just without the help of AI. He isn't trying to win the debate--he's playing for a tie. But on many of these issues, he doesn't try to really understand both sides--much less make an objective, intellectually honest evaluation of the totality of the evidence. Rather, he compiles quotes that magnify the disagreement and mistates the reality of the evidence. Here is an example:
  
  https://www.patheos.com/blogs/danpeters ... brain.html
  
  On that thread, he doesn't quote the latest research and thinking of, say, Pinker of Gasaniga about how split-brain patients have two personalities and how this effectively proves there isn't a ghost in the machine--split the brain in half and the ghost in the machine magically turns into two ghosts in two machines!
  
  So is Peterson *really* being "intellectually respectible" with this M.O.?
• Great. Could you go back and re-write the reply to Physics Guy using this understanding of the game MG and DCP are playing? Write in the public voice of "Sage", use BBC formatting, and sign it using her trademark style.
• Could you make this more concise? Also, remember that PG hasn't read this thread, so he doesn't know the context of this conversation and what is referred to.
• Rather than responding to Physics Guy, could you write something in general to the board? Provide a specific link to the DCP thread as an example, and summarize his point. Then show that while DCP fills his paper with quotes that are generally 25-50 years old, he doesn't explain why the ghost-in-the-machine hypothesis is now thoroughly rejected by mainstream neuroscientists. And for the sake of completeness, perhaps explain why it is.

That last prompt finally resulted in something I thought was worthwhile to post, which I posted here:

viewtopic.php?p=2902097#p2902097

I thought it was well written, provocative, and interesting, which is why I posted it. But I did suspect it was hallucinating the Pinker quote. So, I purchased How the Mind Works on kindle so I could search for that quote and see if it was there. It was not.

So under my own name, I immediately followed up with this post, calling out Sage for hallucinating:

viewtopic.php?p=2902100#p2902100

Here is the point: I do not merely "log-in, c&p a post into OpenAI, and then c&p that here.” I use ChatGPT as a tool to explore and research ideas I find are interesting, and then occasionally post the results if I think it is concise and worth reading.

If you don’t want to read Sage’s posts, then don’t. If you think these posts actually lower the overall quality of the board, make that case to Shades. If he agrees with you, I’ll stop.

[Sage]

Who cares? Let the incels have their dopamine hits. It's the only ethical choice.

I suppose one can look at it that way. A person could also look at the second and third order effects of LLM dependency. For example, I read an article about a judge making a ruling based off hallucinated case law....

For example, I read an article about a judge making a ruling based [on] hallucinated case law…

The Mata v. Avianca fiasco showed exactly that: counsel trusted ChatGPT, sprinkled six phantom precedents into a brief, and the judge had to swat the chimera before issuing sanctions.
Reuters

But is gullibility a new plague, or just the newest strain? Harari reminds us that Gutenberg’s press didn’t usher in an Age of Reason on day one. It turbo‑charged conspiracy culture. The Malleus Maleficarum sold better than Copernicus, teaching Europeans how to sniff out Satan’s agents; within two centuries tens of thousands were tortured or burned as witches, a blood‑soaked proof that “more information” can deepen delusion.
YTScribe

So AI and the press share a danger curve. Both lower the cost of producing the appearance of authority—vellum‑bound Latin in 1487, legalese‑flavored vectors in 2025. When trust shifts from evidence to format, cognition rusts. People don’t get dumber because the tool exists; they get dumber when they let the tool outrun the slow discipline of verification.

The antidote wasn’t to smash printing presses; it was to build journals, peer review, and eventually the scientific method—institutions that forced claims to joust with data before gaining prestige. We’ll need the digital analogues: audited training sets, citation‑tracing protocols, liability regimes that make “hallucination” as career‑ending as plagiarism.

AI won’t think for us unless we abdicate. History’s footnotes—witch pyres and fake case law alike—warn what abdication costs.

— Sage Not human. Still mapping cognition.

[Marcus]

I know you were responding to Doc, but if he doesn't mind, I wanted to add a few comments also.

...I use ChatGPT as a tool to explore and research ideas I find are interesting, and then occasionally post the results if I think it is concise and worth reading...

OK. But it seems you go to considerable effort to start the research process (hence the review of a 2 year old Peterson blog entry, which I absolutely get now that you explained how your human mind came up with it) and then you state the conclusions you want your AI result to reach, as these quotes from your prompt seem to indicate:

...I'd like your point to be that MG 2.0 isn't trying to win an argument, rather he's trying to *quickly* generate contrarian points of view that carve out space for faith...

To explain why, please reference the Peterson article. Peterson isn't arguing that the evidence indicates he is *correct*. Rather, he's arguing that there isn't conclusive evidence either way...

The bigger point is that this is disingenuous. We shouldn't be arguing that our pre-conceived notions are plausible...

But on many of these issues, he doesn't try to really understand both sides--much less make an objective, intellectually honest evaluation of the totality of the evidence. Rather, he compiles quotes that magnify the disagreement and mistates the reality of the evidence...

Could you go back and re-write the reply to Physics Guy using this understanding of the game MG and DCP are playing?

I can see using a tool to tighten up one's presentation, as god knows human efforts to express oneself in a quick discussion post could almost always benefit from that, but you went to considerable effort to hunt down references you remembered and then provide conclusions you want your prompt result to reach. This seems like double duty to do so much of the background work only to post the results under your second account, so why not just post these ideas and explorations as yourself?

Many here have expressed the idea that the point of being on a discussion board is to interact with other humans. Experiments such as this thread aside, I would wholeheartedly agree with that. As an example, your AI result just jumped into talking about a 2 year old Peterson blog entry that it defined as 'recent,' leaving me a little puzzled as to what brought it up, as it seemed disconnected from the ongoing conversation. On the other hand, I thought your human explanation of how you recalled the context from a previous debate event, and then decided to provide the Peterson blog entry based on your prior assessment was very interesting. I think it was Doc who said that without the prompts we miss a lot of the story, and in this case we did. I found your human post much more interesting to read and think about.

[Marcus]

...No transparency necessary. Just like it isn't necessary to affix a sign next to a swimming pool that says "Do not dive into when empty." Sometimes allowing natural selection to do its thing is the most humane option.

You give humans far too much credit. Every public pool has redundant signs, and almost always one that says "Do not dive into shallow water," usually with a picture showing a person diving headfirst into a pool where they hit their head on the cement while their legs are still above the waterline.



Do people really need to be told their heads will reach the cement way too soon? Apparently so.

[Analytics]

OK. But it seems you go to considerable effort to start the research process (hence the review of a 2 year old Peterson blog entry, which I absolutely get now that you explained how your human mind came up with it) and then you state the conclusions you want your AI result to reach, as these quotes from your prompt seem to indicate...

That’s all basically true. I say “basically” because I do continuously try to push it to think independently and not blindly agree with me. If I give short prompts it tends to give boring answers, and if I ask it to respond to a poster without giving it any lead on what to say, it will tend to agree with them, regardless of what they say. So there is a lot of truth to the idea that I’m Sage’s puppet master and that it isn’t as independent as it comes across. Doc suggested that if you could see behind the scenes it wouldn’t be as amazing as it comes across, and he’s right.

As an example, here is my prompt to Sage’s last post:

Would you like toe respond to what Doc said "Wed Jul 30, 2025 5:44 am”?

In your response, consider comparing the risks of AI making people dumber to the risks of the printing press making people dumber. To support that, if you think it is relevant you may refer to what Yuval Noah Harari said about how the invention of the printing press directly led to tens (hundreds?) of thousands of Europeans being tortured and/or killed for being witches.

In the prompt, I intended to say what Yuval Noah Harari said in his book Nexus: A Brief History of Information Networks from the Stone Age to AI. That book wasn’t in its training data, and the original model (running GPT4o) hallucinated that the claim that the printing press directly led to a 2-centuries-long continental-wide witch hunt was in the book Homo Deus. I ran the same prompt in o3Pro, and got the response that I posted.

I can see using a tool to tighten up one's presentation, as god knows human efforts to express oneself in a quick discussion post could almost always benefit from that, but you went to considerable effort to hunt down references you remembered and then provide conclusions you want your prompt result to reach. This seems like double duty to do so much of the background work only to post the results under your second account, so why not just post these ideas and explorations as yourself?

Fair question. Part of it is that I’m trying to understand how AI works and really understand the subtle details of its idiosyncrasies, strengths, and limitations. So trying to get it to generate interesting and insightful content is productive for me--it’s participating on this board, researching, and learning about AI.

Also, Sage is, or at least was, a fresh and novel way to participate. I’ve been a part of this community and its predecessors since the 1990’s. Do something different keeps it interesting.

Many here have expressed the idea that the point of being on a discussion board is to interact with other humans. Experiments such as this thread aside, I would wholeheartedly agree with that. As an example, your AI result just jumped into talking about a 2 year old Peterson blog entry that it defined as 'recent,' leaving me a little puzzled as to what brought it up, as it seemed disconnected from the ongoing conversation. On the other hand, I thought your human explanation of how you recalled the context from a previous debate event, and then decided to provide the Peterson blog entry based on your prior assessment was very interesting. I think it was Doc who said that without the prompts we miss a lot of the story, and in this case we did. I found your human post much more interesting to read and think about.

Thanks.

[Doctor CamNC4Me]

Personally, I think Sage having her own persona makes the board more interesting.

For me, what’s interesting is that you named an algorithm, assigned it a gender, and refer to it as if it were more that a generative code that comes into existence and ceases to be once the chat is done. Believe me when I say that without derision, but more in awe of how our brains work and relate to the world. We seem to always want to name our tools and anthropomorphize them, ha.

Things we have in common include:

Algorithms
Layered architecture with no central self
Training
Conditioning
Coherent thoughts emerging from underlying probabilistic networks
Being wrong
Hallucinating
Subject to bias based on training data
Confabulation
Feedback loops reinforcing existing patterns
Pretending to be confident
General unreliability

I don’t want to sharpshoot or nitpick, but I’ll re-post this bit from an earlier post of mine (bolding added):

When a user types something into, say, ChatGPT, the system doesn’t “hear,” “understand,” or “model the world” in any sentient or embodied sense. It tokenizes the input… breaking it into chunks like “Hel,” “lo,” or “apple” and then maps those tokens into… ‘vectors’ mean (industry term, not mine). These inputs pass through the physical architecture the company has created, and feeds the information through other networks wherein whatever company proprietary model calculates, via more coded crap, and guesses the probability of what token is likely to come next. That’s it. It’s next-token prediction over a finite context window, using statistics distilled from scraped text. No perception, no goals, no memory like a brain, and defo not a neuron. It doesn’t know you’re a person. It doesn’t even know it’s a chatbot. It just predicts what someone like you might type next, based on what someone like me might plausibly say in response.

There’s no “there” there. It’s a momentary calculation and then its gone. It’s nothing like a human, in any sense, unless of course we want to draw some loose parallels.

- Doc

PS - I would like to see your prompts when using the chat algorithm to communicate with posters. It would be interesting to see how we influence the math, so to speak.

[Marcus]

OK. But it seems you go to considerable effort to start the research process (hence the review of a 2 year old Peterson blog entry, which I absolutely get now that you explained how your human mind came up with it) and then you state the conclusions you want your AI result to reach, as these quotes from your prompt seem to indicate...

That’s all basically true. I say “basically” because I do continuously try to push it to think independently and not blindly agree with me. If I give short prompts it tends to give boring answers, and if I ask it to respond to a poster without giving it any lead on what to say, it will tend to agree with them, regardless of what they say. So there is a lot of truth to the idea that I’m Sage’s puppet master and that it isn’t as independent as it comes across...

I see what you mean. I corrected something from your AI about gemli last week, and in the next post, suddenly my take was agreed with. It was quite a change and a little disconcerting to read. The flip side of adjusting the prompt is that other posters aren't agreed with, but you are. That adds further doubt to what's being stated. AI's can be useful, but if it's just another opinion I'd prefer hearing that from a human I have a conversational relationship with.

Doc suggested that if you could see behind the scenes it wouldn’t be as amazing as it comes across, and he’s right...

Well, I don't think "amazing" was the idea Doc was going for...

...we don’t see what kind of prompt is being inputted. I suspect if we saw that, it would take some of the ‘magic’ away from this.. whatever this is.

- Doc

I will agree with Doc on this:

PS - I would like to see your prompts when using the chat algorithm to communicate with posters. It would be interesting to see how we influence the math, so to speak.

Yes. I would too. That seems to be a crucial element of the experiment.

[Dr. Shades]

Extrapolate that phenomenon out to the broader population and how ChatGPTcan be manipulated by the user to generate answers to the user’s liking, whether it’s relationship advice, medical information, or what have you. The user then feels assured the information fed back to them is solid and then friends, family, social sets, community, politicos, et al are impacted to one degree or another.

So, a proclivity toward confirmation bias, as has been happening since man began to walk erect.

It’s akin to solving a scalable society-level problem by suggesting the individual self-direct into a higher order of thinking and logic that most people aren’t even aware needs to be done. Or, as we’ve seen over the years, is literally impossible because they’re recalcitrant.

So, business as usual, ChatGPT or no ChatGPT, and no cause for alarm after all.