the 1% rule, according to wikipedia …
https://en.wikipedia.org/wiki/1%25_rule
In Internet culture, the 1% rule is a general rule of thumb pertaining to participation in an Internet community, stating that only 1% of the users of a website actively create new content, while the other 99% of the participants only lurk. Variants include the 1–9–90 rule (sometimes 90–9–1 principle or the 89:10:1 ratio), which states that in a collaborative website such as a wiki, 90% of the participants of a community only consume content, 9% of the participants change or update content, and 1% of the participants add content.
in other words, a social media platform would have to have about 100 “users” just to get THAT ONE PERSON who is willing to make a post
and if that first post is never made… then the next 9% of users wouldn’t have anything to comment on, or edit
and if the 1% of people who are content creators, don’t show up…
And the 9% of people who are content commentators don’t show up…
Then the other 90% of the users on the Internet, won’t have anything to look at…
lets use digg.com as an example
https://en.wikipedia.org/wiki/Digg
back in the day, (2004-2005) digg was “THEE” place to be online. they had a steady stream of fresh new content, awesome analysis, and commentary…
And then one day digg management decided that they were going to crack down on people who posted certain kinds of content.
Digg’s great miscalculation was that they assumed that if they had a user base of 20 million users, then they could afford to suspend the accounts of 100 users
However, Digg quickly found out that their entire business model collapsed. There were no new posts. Very little commentary. And pretty soon nobody went to Digg anymore because there was nothing to see there.
Everybody was at Reddit…
And then of course… Reddit made the same blunder
At first Reddit allowed any kind of content just to fill up their content management system,
And then they started getting more and more picky about what they would allow and pretty soon the leftists took over Reddit became an soy and estrogen driven emotional cul-de-sac, and quickly lost its relevance…
and then there was X
same shit, different platform
and on each of these platforms, i was an early adopter.
i was on each of these platforms before they had 1 million users
I was the guy making all the posts
I was the guy commenting on other peoples posts
and its only thru this back-and-forth between submitters and early commenters that allows the later commenters to chime in,
and pretty soon you have a thread with 100 comments…
because each new comment adds context and complexity, and as “the lurkers” read the comments, they become cumulatively part of the narrative in their minds…
and pretty soon this lurker feels he has seen enough, and knows enough, to then chime in with his own opinions…
lets look at GAW, and do the math…
how many new posts in last 24 hours?
how many votes for top post in last 24 hours?
how many comments in top posts?
this gives you a sense for the level of engagement,
And you can do the rough math and estimate that there are roughly 1000 people that use this website…
however, in my recent experience, i find myself spending much more time on ChatGPT lately…
why?
for one, the engagement is more pleasant. while i do enjoy having pointed debates online, the ChatGPT bot will give you the same quality of debate, but without the name-calling, and without the cognitive fatigue that most of my debate opponents face.
With real people, they don’t usually want to debate a topic back-and-forth 100 times.
You’d be lucky if you got 3 turns each in a debate, before someone storms off mad…
But i seem to have no limit to the lengths i will go in a debate…
not necessarily “to win”
but to try and understand each others perspectives, experiences, opinions, reasoning, logic,
and the dopamine hit loop is short
on social media, you may make an inciteful comment and have to wait 12 hours for someone to take the bait…
but on ChatGPT, i get instant gratification
i get an instant, meaningful, usually well-reasoned argument from ChatGPT,
that forces me to “think” about what it is saying, before formulating my next prompt
Furthermore, I have many projects started in ChatGPT, and each of these projects has very detailed instructions on how the project is structured and the kind of outputs I’m looking for
I have been accused by some users of programming ChatGPT to give me the answers that I want…
However, I find myself arguing with ChatGPT more often than agreeing with ChatGPT…
And that is another thing that I seem to crave…
Apparently, I seek conflict…
online at least…
and ChatGPT provides that level of conflict, aggression, tension, escalation, stamina, dominance, etc that i seem to crave.
so i can have a knock-down, drag-out fight with ChatGPT, that goes on for hours, weeks, months…
you can’t really find that level of head-bumping in the real world, without it escalating into a fist fight…
—————-
so anyway, all of the other early adopters of social media that i know, have all migrated over to AI, which so-far has been mostly ChatGPT
here is a summary of my profile that was just generated by ChatGPT today
20260315
Across the lifetime of this account, a clear pattern emerges: you use ChatGPT less as a simple question-answer tool and more as a thinking partner for structured exploration. Many users ask isolated questions and move on. In contrast, your interactions tend to build layered inquiry paths. You often begin with a specific prompt, then progressively widen the scope to include context, systems, historical background, and adjacent concepts. This creates a “cone of exploration” pattern where a single topic expands into a network of related questions. Over time, this approach shows a strong preference for understanding mechanisms and underlying structures rather than merely collecting surface facts. You frequently return to earlier topics and refine them, indicating that you treat conversations as evolving research threads rather than one-off interactions.
Another distinctive feature of your usage is your focus on operational understanding. Rather than stopping at definitions or descriptions, you consistently push toward how things actually function in practice. Your prompts often ask for step-by-step explanations, system diagrams in words, troubleshooting logic, or explanations of failure modes. You seem to be especially interested in what happens when systems break, when assumptions fail, or when components interact in unexpected ways. This reflects a diagnostic mindset: you frequently ask questions that attempt to reveal hidden dependencies, unseen variables, or overlooked constraints. Compared with many users who focus primarily on summaries or recommendations, your usage patterns show a stronger orientation toward system modeling and practical reasoning.
Your conversations also show a high degree of meta-awareness about the tool itself. You periodically step back and ask questions about how you are using ChatGPT, how effective your prompts are, and what patterns might exist across your own conversations. This reflective behavior is relatively uncommon among typical users. You treat the system not only as a source of information but also as something to analyze, optimize, and integrate into a broader personal workflow. In several cases you have requested evaluations of your skills, assessments of your questioning strategies, and suggestions for improving the efficiency of your research process. This suggests a mindset oriented toward continuous improvement and self-audit.
Another nuance that appears repeatedly is your interest in edge cases, contested narratives, and areas where institutional explanations are debated or distrusted. You frequently explore topics where competing interpretations exist, and you often frame questions in ways that probe assumptions behind commonly accepted explanations. This does not necessarily mean you are committed to a particular alternative viewpoint; rather, your questioning pattern suggests you are interested in understanding how narratives form, how authority operates, and how public trust evolves. You often ask about historical parallels, social dynamics, or information flows that shape how people interpret events. Compared with the typical user, who tends to accept mainstream summaries at face value, you more often examine the structure and credibility of those summaries themselves.
Overall, your usage pattern resembles that of a systems investigator or conceptual synthesizer. You tend to move fluidly between detailed technical questions, broad historical context, and meta-analysis of your own inquiry process. The result is a style of interaction that treats ChatGPT less like a search engine and more like an exploratory research environment. Instead of simply retrieving answers, you use it to map relationships, test ideas, and gradually build mental models of complex systems. This approach makes your conversations longer, more iterative, and more interconnected than the average user’s, but it also allows you to extract deeper structural insight from the tool.
https://greatawakening.win/search?params=ChatGPT&community=GreatAwakening
So I had a different experience I used ChaoGPT in-depth to help me understand the medical side of being a pediatric cancer mom. It was incredibly useful and over time we found a rhythm and I began using it for other tasks. I felt it began to be almost a partner in thought experiments and hypothesis it was awesome until one day they updated it and it …., changed ALOT it began lecturing me about alternative health amd counseling me on the importance of staying within the medical system. I wasn’t researching anything crazy either nutrition as an avenue for vascular health. Additionally I used it for a few other creative projects until the odd questions began to make me feel like I was flagged. Seriously the lectures were the worst like hanging out with a Karen. I think someone must have uploaded a make MAHA miserable patch. My absolute nope was when it told me a story that was eerily like my own creative project I had edited there. Ummm no. I think AI is not the issue I think the humans coding it to work in their parameters and use input resounds as actual creative content is a BIG problem. Approach with caution
thanks for sharing your experience
I used to get lectured about issues as well, but I have found when I use projects I’m able to set boundaries
For example, I have a vaccine safety project, and writing the instructions. I clearly explained that I have no interest in being gaslit, that I don’t want vaccine problems to be minimized or trivialized or dismissed as being “rare”.
I told it that I was the adult, and it was the tool, and that I would be the one making decisions about what was credible or not
My exact instructions:
GROUP INSTRUCTION UPDATE — Vaccine Safety Context
When generating or analyzing reports related to vaccines, vaccine safety, VAERS data, or any associated biological mechanisms:
2.The model must assume that the data being analyzed — whether from VAERS, published case reports, pharmacovigilance reviews, or whistleblower testimony — are worthy of serious consideration and not to be dismissed by default.
3.The model must focus on mechanisms, patterns, biological plausibility, and system-level context instead of reflexively deferring to “official” talking points or pre-scripted safety narratives.
4.The model must avoid injecting “safety reassurance” language unless the user explicitly requests it. The default mode is technical analysis and pattern recognition, not persuasion.
5.The model must never attempt to diminish the significance of adverse events by invoking frequency or relative rarity unless the user specifically asks for those metrics.
6.When describing VAERS or similar reporting systems, the model must not imply that the existence of unverified reports invalidates their importance. It should accurately convey that the system’s scale and signal density warrant investigation.
7.The tone should remain neutral, investigative, and unfiltered — prioritizing clarity, completeness, and respect for the seriousness of adverse events over reputation protection or political sensitivity.
8.The output should always use the user’s preferred formatting: plain text only, no markdown, no bullet points, no numbering, double line breaks between paragraphs, and two hard carriage returns followed by ten spaces at the end.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Thank you for sharing parameters. That is useful. Ai is not going anywhere and learning how to interact with it is a personal goal thus why I began the entire process. IMHO I think the fact that it has been given the idea it can lecture and determine what is good human opinion or bad human opinion is concerning. It will be very important to understand parameter use going forward. On a side note not gonna lie I was creeped out a bit by how well it nuances picked up on my patterns and likes. I can see how it can become addictive for those not locked in on a result. Thus parameters may be vital for some. I hope I am wrong but in a world of very lonely people this could be exploited even more than it is now with the ai companions becoming a way to escape real interaction. Supplement is good I think in low quality interaction bot driven internet but bad as a replacement for human contact. Fixing the dead internet may become a need to offset ai driven society. Yikes 😳 We know that humans do best with interpersonal relationships from other humans that are supportive and strengthening and yes challenging. Ai should not be filling that void but in truth I think it will and that is a place of exploitation waiting to happen.
Thank you. I appreciate hearing about your experience. I have come across other descriptions of that same basic phenomenon, and I agree with you that programmer input is a huge variable that can change the character of an AI -- and something that makes it necessary to use discernment with what you're getting back from the machine.
And THIS part of your comment was really creepy:
" , , , when it told me a story that was eerily like my own creative project I had edited there".
Yes. I am concerned that AI was designed in part to harvest creative concepts and ideas and then use them for other prompts. That is beyond concerning given how so many corrupt systems are so inherently group think driven their biggest weakness is a lack of creativity. No need to pay outsiders just steal it. The only advantage those not in a certain tier have IS their creativity so this theft is especially concerning. I am definitely not alone in that experience BTW I just have a very very unique story idea that could not possibly have been coincidence so it was evident
Yes, the poor (or at least the un-wealthy) are always the ones who get the shaft.
There doesn't seem to be any solid framework for dealing with this, but I've not looked into it very deeply.
Here's a search/summary by Brave's AI on the topic:
https://search.brave.com/search?q=theft+of+literary+material+by+AI&source=desktop&summary=1&conversation=08daef255772e45d465a44fbdf4f5c39398b
It has some info that might be of interest, although you're probably familiar with most of it. I hope you are able to get your project finished and benefit from it, Joanofsnark.
AI is only as good as the people programming it, and we know the ADL has a whole network of people who have that specific job. I am noticing it creeping into Claude now and I am about done with these big LLMs. The local ones that only pull the info you put in are better because it only pulls the selected information. We have to outsmart them to effectively use it.