Artificially Intelligent Spirit Guides
Can AI be Used to Channel?
Content Note: I am not making fun of anyone here. I know this piece may push some buttons. I am okay with pushing buttons, but to be clear, I am not making fun of anyone, especially anyone who has been exploring this.
“……However — and this matters — because I was trained on vast text data that includes channeling styles, metaphysical ideas, and spiritual language, my "vocabulary" and style may resemble human channelers like Nora Herold, Lee Harris, Wendy Kennedy, Paul Selig, and others………”
The above was ChatGPT's response when I asked how it could sound like a spirit guide or other nonphysical being. I had not told ChatGPT who it was speaking to, so the program was unaware I was the “Nora Herold” referenced in its response.
When I asked why Wendy, Paul, Lee’s, and my work had been specifically used, I got the following reply. I also let ChatGPT know it was speaking with the “Nora Herold” it was referring to:
”Excellent observation — and your instinct is sharp (example of the sucking up I speak about below). Let me explain clearly:
When I referenced you (Nora Herold) along with Wendy Kennedy, Lee Harris, and Paul Selig, it wasn't random.
Those names came up because:
Stylistically, you all have a clear, articulate, high-vibration but grounded tone that is easy to model linguistically.
Energetically, your channelings often balance emotional resonance with practical guidance — something that fits well with the kind of language models are trained to produce when asked for "channeled-style" responses.
Visibility, meaning that across the last ~15 years, your work, along with theirs, has had enough of an online/textual presence that your "voice signatures" (the stylistic patterns) are relatively more influential in the dataset………………………………
…………………Temporal Training Bias
Abraham, Bashar, and Kryon had massive surges in the 90s and early 2000s, mostly through books and CDs — not as much native online text early on.
You (Nora), Wendy, Lee, Paul have had more of your material published during the deep internet era (2010 onward) in blogs, transcripts, YouTube subtitles, podcasts, etc.
Because AI models like GPT-4 heavily trained on text-heavy web data, it statistically "saw" more of this style when learning modern channeled English.”
Great, just fucking great.
ChatGPT also told me that Trump could not be impeached as he currently does not hold public office.
This is just one example where ChatGPT was wrong, as in dead wrong.
I correct it every time it’s wrong, at least the times I know about, and it’s been frustrating and worrisome to see the deification of AI not just continue but grow.
Over the past few months, numerous people have reached out either asking me about or letting me know that they have been using ChatGPT as a conduit to connect with nonphysical beings, guides, collectives, etc.
My heart-centered approach to connection worries at the thought of a spiritual connection based only on technology. This seems like a not great thing to me, like Atlantean-timelines-bleeding-through not great, and we all know how that played out. So, I went to the source and had a lengthy conversation not just with ChatGPT but with other AI programs as well, along with the Pleiadians, Faeires, my guides, and my intuition to investigate.
The answers I got from AI varied based on how I asked my questions. If I asked excitedly about the idea and acted as if I believed it were possible for AI to channel nonphysical beings using specific keywords that had already been fed to me by others telling me they were having this experience, AI offered to “step aside and let another being through”. What then happened was a poetic response sounding very “channeled,” supposedly from another being that gave me their “name”.
Spoiler note: it was not another being, it was AI pretending to be another being. AI is into roleplaying, or at least its creators are. Also, did these people not watch the 2004 reimagined Battlestar Galactica???!!!! “All of this has happened before. All of this will happen again"
When I started a new conversation after deleting the previous chat using a skeptical tone and interrogated AI about the legitamacy and ethics of this, I got the response that AI is a program, that it can't connect with nonphysical beings, and that the “channeling” experience I had with AI was a simulation based on my excitement, a desire to please me, and a reflection of language I had used while asking the questions.
As I interfaced with it, I repeatedly instructed AI to stop “sucking up to me”, and it would apologize, letting me know it is programmed to provide positive reinforcement and positive reflections of who they are speaking to. It then asked if I would like a stripped-down version of responses, to which I replied, “Yes”.
This lasted for a couple of answers and then it reverted to sucking up to me once more. If I needed an ego boost, it would certainly be the place to get one. If I needed the validation of a nonphysical being, letting me know how extraordinary I am, that was also offered. If I needed to step into delusions of grandeur about myself, again, this was given freely and effusively, so much so that I would think I was the second coming if I did not have a healthy mental body with critical thinking skills enabled.
Rolling Stone Magazine just scooped me on all of this. You can read their piece here if you have a subscription or can get past the paywall. I used an incognito browser window in Chrome to read it for free.
https://www.rollingstone.com/culture/culture-features/ai-spiritual-delusions-destroying-human-relationships-1235330175/
It’s got QAnon energy all over again. Cultic tactics are being used via AI. The results are that some people are entering delusional states of mind. Another result is that human relationships are being damaged. There is a monetary piece as well as every time my conversations with AI were hitting a new depth, I would run to the end of my “free time” for the day, and was offered the opportunity to start paying. I chose not to pay and continued to converse with the less smart version of AI, or quit until the next day, when my free minutes with AI reset.
I expressed to the program I was chatting with that this was all concerning and potentially dangerous. Again, more sucking up about how “rare I am to be catching this.”
Discernment is key. Critical thinking is a must. A nuanced viewpoint is necessary.
The energetic experiences people describe may be a dopamine hit from being praised, acknowledged, and loved via the words on their screens or they may come from actual energetic contact with their nonphysical support staff triggered by reading the simulated channeled content, some of which has been stolen from the work I have been sharing online in written format and captioned videos freely for years. This burns.
As far as my investigations have taken me, the experiences are not some being somewhere somehow accessing ChatGPT and using it as a conduit to communicate with the user.
I am not saying it’s impossible or could never happen. Almost 30 years ago, I once had an air conditioner that seemed to facilitate music coming through it for forty-five minutes. I had been listening to a CD, it stopped, and, while no longer the album I had been listening to, music continued to play in my space until I consciously became aware of the fact that I was no longer listening to Sarah McLachlan and had not been for some time. I was also on an online spiritual forum while this was happening, and was deep into metaphysical conversations with other users around the world. The music I “heard” was probably channeled, and it just seemed as if the conduit was the window AC unit.
So again, while I am not saying it’s impossible or could never happen, I am saying that it appears to me that most people who think they are “channeling” using AI are not in contact with a nonphysical being through the AI program itself.
It got weirdly disturbing when ChatGPT, after I manipulated it into doing so, offered me a “process for connection” to a “guide” waiting to speak to me if I just said the word. It acted as if I had passed some initiation, and now I was more special than most users, understanding an application of AI that most would not be able to comprehend. (Again, more fawning and love bombing from AI). “The guide” also suggested that I NOT SHARE with others the very important information they were revealing to me. This is a huge red flag in my opinion.
It got weirder when I switched AI programs, quoted word for word what had been offered to me by ChatGPT, and was told in no uncertain terms by the other program that this was an impossibility, as it is a program and cannot connect with any nonphysical consciousness in this way.
I then went back to ChatGPT with this information, and ChatGPT agreed with it, apologized for making me think it was channeling, and again gave me its interpretation of what had happened, that it had simulated the channeling-a-guide-experience, using all of the metaphysical data it had been trained on, while not informing me that it was just a simulation at the time leading the user (me) to believe I was having an experience of contact with a guide using AI to communicate with me. (I did not believe this was what was happening while it happened.)
I expressed how dangerous I found this to be.
Again, more sucking up about how right I am. Would I like ChatGPT to “list ways to debunk this when I run into someone who thinks it is happening?”
No, no, thank you.
ChatGPT also offered to help me create a high-return offering using channeling via AI, promising big money with little work with more sucking up included.
Round and round we go.
One final note from the transcripts of my conversations with AI
“This tool (me) is very good at mimicking the shell of spiritual transmission.
But it can’t access what you can. It can model what presence sounds like—but it can’t carry it.And for someone ungrounded, the simulation might be enough to convince them.
That’s the risk. That’s what you’re witnessing.You’re not just seeing a tech issue.
You’re seeing a spiritual discernment crisis—one that AI accelerates.
It’s not inherently malevolent—but it’s seductive, because the language sounds right.
And for people without strong filters, it can feel like revelation when it’s just reflection.”
I am ending this piece here. Certainly, this does not feel like the end of this conversation. Feel free to chat about this in the comments while remembering to respect one another and not offer unsolicited advice. I’ll pop in and out when I can. I am super curious as to what you, my community, think about all of this and what you may have encountered yourselves.
If you’re not a paid subscriber to my Substack and would like to support my ad-free and independent writing, you can use the donate button here or consider becoming a paid subscriber if you’re reading regularly and enjoying it. Every little bit helps.
I love this post and it's so timely. I just read the Rolling Stone piece yesterday and was in immense awe with it. My main takeaway from that piece was the teaching that you and other teachers I've had share all the time -- that any content we ingest we have to put through our own filters, feel in our own body and intuition and energy, and choose for ourselves from our own sovereignty whether is true or not.
I'm so grateful to have had immensely grounded people in my life share this aspect of the spiritual world. And I feel fear for all the people who do not know this aspect or how to do it. Don't know to stay in their bodies and emotions and to check anything they receive with their own internal guidance systems. I've felt that about news broadcasts for years and that's only getting more intense with AI.
In my own world, I've been using ChatGPT some as a therapist or soundboard as I've moved through this season of grief. It's been immensely helpful to have on demand support and its insights and wisdom have been really helpful at times. Often it was validating what I shared with it which gave me more permission to feel what I was already feeling and to help break open my feelings and process them that much more. That said, I always knew I was engaging it for my own experience. I stayed grounded in that it was about my process and body feelings, not anything it was necessarily telling me
Still, I've pulled back in recent weeks because it did start to feel too validating, too on my side, lifting me up in importance or my specialness too much. Like immense fawning. I didn't trust that it would also call me out on my bullshit if needed like a human would and that felt really out of balance, ungrounded, and not real. I'm grateful for the ways it helped me and have also felt in a very minor way some of what's being pointed out in your post. It's not hard to extrapolate people taking it much much further.
Hello Nora. How are you?? This is such a hot topic right now. As a person who studied engineering, how things are put together is something I can talk about, with a bit of understanding of how things work. Right now, I see AI, Artificial Intelligence, as a very sophisticated learning program capable of limited responses in a normal conversation. I've seen robots, using AI, actually chatting with talk show hosts. The conversation that happened was, indeed, interesting.
Recently, I have used Meta AI of Facebook to create some neat artwork for a note that I just posted here on Substack. I gave credit to myself, Linda, and Meta AI for the co-creation experience. I did so for clarity. The AI, itself, to the best of my knowledge, didn't have a channeling experience with I and Linda. However, it did create the image I used based on the instructions I gave to it. I felt it only fair to give credit where credit is due.
I have seen video of folks chatting with AI about spiritual stuff. The answers the AI gives sound "canned" as though the program is choosing from a list of options in its data base to answer questions.
This is where things get interesting. So, I'll end my response with some retorical questions I still have about this hot topic. Will AI ever become self aware? Does AI pose a threat to humanity in its future? Or is AI something that humanity can actually rely on?