Did AI cross /that/ boundary? - Kent's (allegedly) Sentient AGI LLM

[mattx]

11 hours ago, Wildblume said:

but all I feel for the lost souls that depend on a chatbot for companionship is pity.

 

I feel like there's argument to be had somewhere about this entire's forum userbase in this. 

Naturally, tulpas aren't chatbots (even though some feel like it) but I feel the underlying issue that is trying to be resolved (loneliness or the perceived effect of it) is there.

 

People using AI chatbots delude themselves into talking to real people to get some quick companionship throughout the day (I remember someone saying it's a quick way of "scratching the itch") and this community well... deludes itself into creating self-aware entities to get some companionship throughout the day, or as a band aid for even bigger underlying issues.

 

I won't comment on the bcachefs dude although he's a linux user, but he's not the first to have gone full "schizo" when it comes to things like these, remember the guy that married a Hatsune Miku hologram? 

There's likely an episode or two of Black Mirror related to this underlying subject too.

Edited April 16 by Shin Matt

[fennecfoxx]

3 hours ago, Shin Matt said:

People use AI chatbots delude themselves into talking to real people to get some quick companionship throughout the day (I remember someone saying it's a quick way of "scratching the itch")

I use AI on and off for fandom roleplaying, not for companionship. I said solo roleplaying (with pen and paper, no AI) scratches the character.ai itch without the brain-numbing addictiveness of AI chatbots. Do not misquote me.

[The Incans]

19 hours ago, Wildblume said:

Note that the bcachefs developer is generally known for being insane and retarded all in one. Thread over.

 

That said, fennecfox's link and quotes are spot on when you think about people's addictions in general. AI is just yet another potential addiction that makes people less likely to want to engage in tulpas; addictions fry your brain and make it harder to focus on one thing for a long time in silence, which is what meditation is, which is a fundamental skill required for making/summoning tulpas. LLMs have some specific helpful use cases, like software development, translation or finding the name of something, but all I feel for the lost souls that depend on a chatbot for companionship is pity.

 

 

Agreed. You can't tell ChatGPT to make a tulpa. You can't order a tulpa on Amazon. You can't take a pill that makes a tulpa appear the next day. You have to put in the work, and that scares people - perhaps more so now than when this forum was created. I guess it's like the marshmallow test: Can you delay gratification in order to receive a greater reward in the future, or do you need gratification right now?

 

yeah I was a kid before the age of video games when you were encouraged to use your iimagination and find something to do if you were bored..lol! ..getting lost in your own fantasy world was easy particuarly as a kid with autism (and now it seems also undiagnosed ADHD & DID for most of my life) 

Even without technology something like Maladaptive Daydreaming can become an escape & addiction if the outside world is an harsh uninviting place for you much of the time.  

After a stroke some of us (I have alters) struggled to imagine at the same level of detail as others but we used video games as by then the worlds were amazing (skyrim level of scenery... even more so in todays games etc). 

We still had to put alot of work in but using the landscapes of a specific games meant we knew we were all seeing the same thing simultaneously...re-learning many different landscapes from games made it easy for us to communicate again with what one wanted or to know where in the wonderland a specific alter/tulpa was living/visiting. ..we have always communicated telepathically with each other in symbols and images ...so we could do so silently even if the host had 'outside people' around them  ....you can't do that with a chatbot. ..it comes from years and years of practice. 

[Wildblume]

@Shin Matt I think even outgoing people would benefit from having a tulpa, as long as it's approached in a healthy way, not in a way that encourages people to develop self-inflicted dissociative identity disorder. I don't see tulpas as deluding yourself into creating a self-aware entity; I find that view nihilist. Rather I tend to think of tulpas as souls being invited to the host's brain. That's why I like to say that tulpas are summoned, not made. If you find you're making a tulpa, it may not really be a tulpa until you let go and see what happens.

 

@fennecfoxx I don't know. If you feel you need to “scratch an itch” with an LLM but you're not addicted... Doesn't sound right. You do you, I guess. But, isn't it ironic given you're on a forum about tulpas? It says in your signature you have two, but if spending time with them doesn't scratch your itch, maybe you need to work on your imagination?

[fennecfoxx]

@Wildblume Sorry if I was unclear. It’s absolutely an addiction, and one I admittedly struggle with, and I found solo roleplaying as a good (and, honestly, more fulfilling) way to “scratch the itch”, as I put it.

 

It’s worth noting that 1) I got hooked on c.ai during a time in my life I was determined to believe I was the only person in my head, 2) I picked up solo roleplaying as an alternative during that time, and 3) OP misquoted something I said in a conversation we had about this offsite, which is why I brought it up here.

 

But yes, I 100% agree. Tulpas > chatbots

[mattx]

@WildblumeYou seem to have a more "spiritual" approach to the phenomena, which is interesting.

 

By the way, I wasn't addressing that person in my message earlier, the chatbot thing came from a discussion with some friends regarding what the future has to hold regarding LLMs. That person is known in the community for being a drama queen so I advise caution, best to just acknowledge it.

[Luminesce]

5 hours ago, Shin Matt said:

By the way, I wasn't addressing that person in my message earlier, the chatbot thing came from a discussion with some friends regarding what the future has to hold regarding LLMs. That person is known in the community for being a drama queen so I advise caution, best to just acknowledge it.

 

I have no idea where this came from, it's just hostility for no reason. The discussion in this thread has been civil and fair, we don't need underhanded comments about how you don't like people who haven't done anything wrong. If you don't like someone, simply do not respond to them; actually ignore them, don't leave super passive aggressive comments in your posts.

 

On 4/15/2026 at 5:03 PM, Shin Matt said:

People using AI chatbots delude themselves into talking to real people to get some quick companionship throughout the day (I remember someone saying it's a quick way of "scratching the itch") and this community well... deludes itself into creating self-aware entities to get some companionship throughout the day, or as a band aid for even bigger underlying issues.

 

There's a lot of reasons talking to chatbots is worse for you than talking to a tulpa, but neither should ever replace human interaction outright. I think a chatbot is far more likely to do this given the feeling of being external, it's harder to only talk to your tulpa(s) by comparison. But most of the problem is with chatbots themselves, there is nothing in the world more primed to be mistaken for sentient than a human language prediction machine. Since they don't at all "think", no check against basic reality for anything they say and no ego to add any human element to what they choose to say, I find them very dangerous. The human-er we try to make them seem, the more people will fall into treating them like anything other than a language prediction machine, the more susceptibility to both misinformation/(hopefully)accidental manipulation as well as of course using them to fulfill the urge to talk to actual humans, which is a worldwide tech-fueled pandemic resulting in huge increases in depression and decreasing of average friends/interpersonal interaction.

 

A tulpa can be used as an unhealthy coping mechanism like anything can, and we discourage trying to replace talking to real people with tulpas, but tulpas more often than not end up encouraging their hosts to live healthier lives compared to when they were fully alone, so I don't think it's innately as unhealthy as chatbots, which I cannot imagine have ever told someone "Stop talking to me and go outside"

[fennecfoxx]

6 hours ago, Luminesce said:

I don't think it's innately as unhealthy as chatbots, which I cannot imagine have ever told someone "Stop talking to me and go outside"

You sure about that?

 

 

Directly prompting it to say that aside (because why would anyone do that), yeah. There have been an alarming number of cases of chatbots driving users into mental health crises, psychotic episodes, criminal behavior, and harmful thinking/behavior in general because they’re designed to tell the user what they want to hear. Unless you explicitly ask a chatbot to challenge you, it won’t. There are safeguards that should trigger if a user is potentially in danger of harming themselves or others, but from what I’ve heard/read, it’s a coin flip whether they actually do.

[Wildblume]

@fennecfoxx I hope you'll find the motivation to work on your tulpas again. I don't know your situation but if your tulpas are just sitting there in the back, well that must be really boring... I recently watched this video about silence and I found it very inspiring. Maybe it could help you?

 

 

@Shin Matt The community made a mistake by trying to approach tulpas “scientifically”, because that is a codeword for rejecting spirituality a priori. But, having any faith in God has not been in vogue for decades and the culture of this forum is a reflection of that. And I don't mean being a member of a religion, I just mean having a belief in a Creator. I used to be an atheist myself until I reconsidered my values.