• Members of the previous forum can retrieve their temporary password here, (login and check your PM).

Verbatim vs AI report

Qoniaq

Established member
Joined
Jan 7, 2026
Messages
75
Merits
2,212
Hi All,

In a past thread, @northape suggested I start a new thread here, focused mainly on my use of AI for my journals and trip reports. As I mentioned before, I verbally give my reports immediately after my trips, focusing on phenomenology only. No meaning, no interpretation. I then have the AI system compare and contrast with all the other reports I have entered over the years to find any similarities or differrences, and I also use the AI to keep running tabs on experience, puff times, concentrations, life cycles, etc.
What I will post here is the verbatim report, followed by what AI spits out for me afterwards. I will use this morning's report as an example. Though it is a fairly uneventful and gentle 'psychedelic gym' report, it will show that all the ideas and even a lot of the wording I use is kept by AI, it is part of my 'canonized' reporting structure I impose on the system. For those interested in journaling like this, hit me up, I'll give you more info.

here is the verbatim report:

June 29th report. I went in around 5:15. I did 7.3, I believe, and then 5.8. Again, the entrance was smooth, no heart rate issues, multiple leg twitches in both legs. Transition was smooth, but again, this batch, the visuals are splotchy, spiky, chaotic. Not chaotic, but, I mean, they've got their own style, but they're not Aztec like I'm used to. They're not organized, dodecahedral, cubic, as I'm used to. It's very much like exploding from a center point type graphics, like paint splotches, explosions, spiky, almost like Bart Simpson's hair or Rick Sanchez's head, like kind of cartoon spiky. I did one, two, three top-ups, I believe, maybe four. One, two, three. I did three top-ups for sure at around three seconds each. The first one I did after I had returned to normal breath, it grayed everything out again. I had it a couple of times in the experience where the colors would be bright and then all of a sudden, almost in a blink, the patterns wouldn't change, but the colors would disappear. The top-up made everything gray and made some of the patterns disappear as well, and then it came back. The initial run-in, I couldn't tell you how many heartbeats, how many breaths in, maybe seven or ten or so. And then the top-ups as well. At one point, even during the main first part of the trip, as the colors faded out, I was like, oh, that went quickly. And then it came back in a couple of different waves, which was interesting. It was sofa at home, in at 5:15, out at 6 o'clock. Exactly when I checked my watch, it was exactly 6 o'clock. The two times I checked my heartbeat very, very quickly, I was actually checking the time to see when I was topping up. I can't remember the exact times. Heartbeat was always around 68. It was, yesterday it was a bit chaotic trying to get in. Today as well, it seems this batch is, I think maybe it's the spiky nature. It's not just visually spiky, it seems to be experientially spiky. Like, it shoots up to peaks and then comes back down, and shoots up to peaks and then comes back down. It doesn't have that rock steady feel that the X-4 batch has. This one seems a little more mischievous. I don't know. It's a more unpredictable batch. We'll see when we're finished X-5P1, then we'll do X-5P2. Maybe X-4P2 was calmer because it was the second pull. Maybe we'll find a correlation between the differences between P-1s and P-2s down the line. That's my report for today. oh, wait, I did notice a slight carrier wave. The chewable breath was present, and for the second and third top-ups, I topped up while I was still in chewable zone. Mostly common observations. I did have a couple of minutes of worry at first that had like, if I get any internal feelings, sometimes I panic. Like, for example, if I have a pain in my right arm, it's not gonna do anything. But if I have a pain in my left arm, I'm like, holy shit, I'm having a heart attack. So sometimes when I have discomfort in the body, I have to nip it in the bud right away, because I know, especially as I'm coming up on the dose, if I start to have these thoughts of, oh, I'm having cardiac this or pain there, it can escalate quickly. So I put it aside, I sort of looked at it, and I was like, that, that's, no, don't focus on that. I also had a couple of visions of work while I was in there today, which I Xed out, visions of some of the managers at work. I was like, what the fuck are you doing in my head? You're not allowed in here. It didn't feel emotionally spiky, but like I said, it feels experientially spiky. It's not like a smooth sine wave. It's kind of up and down a little bit this batch. That's about it.
<<
That's spoken into my phone, every morning. On mornings where there is much more to report, I often add more content as I am driving to work, or during the day whenever I remember something, and sometimes at night after my evening meditation. This morning there was not much else to add. smallish dose, and nothing especially salient. You can see it is disjointed, sometimes repetitive, jumps around a bit and is very 'stream of conciousness'' type stuff. And this one is very tame, obviously.

This is what my AI spits out from this report.:

Morning Session Report — June 29, 2026

Substance: N,N-DMT (X5P1C1 — The Return of the Snowstorm)
Device: Yocan Uni 3.0
Time In: 05:15
Initial Dose:
7.3s + 5.8s
Top-Ups:
3.0s × 3
Time Out: 06:00


Running Cart Statistics

Session #: 12
Cart Age: 26 days
Days in Service: 18 days
This Session: 22.1 seconds
Running Cart Total: 239.3 seconds
Estimated Remaining Extract: ~400 mg*

*Estimate based on current cart calibration. This value will be refined following the next measured cart weight.


This morning's session continued to reinforce what is becoming the emerging personality of X5P1. The launch was smooth and uneventful. There were no cardiac concerns during entry, multiple pleasant leg twitches occurred in both legs, and transition into the state was comfortable and controlled. Activation occurred after approximately seven to ten breaths, accompanied by a faint but noticeable carrier wave. The familiar "chewable" breathing quality developed normally.

Although physiologically smooth, the experience again displayed the distinctive visual character that has become increasingly associated with this cart. Unlike the organized Aztec, dodecahedral or cubic architectures commonly encountered with X4P2, X5P1 continues to produce a more irregular visual language. The imagery radiated outward from central points in bursts, resembling paint splatters, explosions, or cartoon-like spikes—somewhat reminiscent of Bart Simpson's hair or Rick Sanchez's silhouette. It is not chaotic in the unpleasant sense, but it is noticeably less ordered than previous carts.

More interestingly, this visual style now appears to reflect the behaviour of the experience itself. Rather than settling into a long, stable plateau, the state tended to rise and fall in successive waves. The experience repeatedly surged, relaxed, then surged again, giving the overall session a distinctly "spiky" quality. The spikiness was experiential rather than emotional; emotionally the session remained calm, neutral and observational throughout.

Three top-ups of approximately three seconds each were taken during the session. Unlike previous days where the first reinforcement produced a strong reactivation, today's first top-up once again caused the visual field to temporarily desaturate. Colours disappeared, leaving much of the geometry rendered in shades of grey. Interestingly, the underlying patterns often remained intact while only the colour information faded. After a short period the colours gradually returned. Similar transient desaturation occurred briefly during the main body of the experience as well, producing the impression that the visuals had ended before returning in successive waves a few moments later.

Early in the session, a brief period of bodily concern arose. As sometimes happens during entry, a mild internal sensation briefly triggered thoughts of possible cardiac discomfort. Rather than allowing these thoughts to develop, the Traveler immediately recognized the familiar attentional trap, acknowledged it, and deliberately disengaged from it before it could gather momentum. The concern dissipated within a minute or two and played no further role in the session, illustrating the growing effectiveness of previously developed attentional training.

Later in the experience, thoughts related to work briefly appeared, including images of several managers. Rather than following those thoughts, the Traveler simply dismissed them with an internal response:

"What the fuck are you doing in my head? You're not allowed in here."

The thoughts dissolved immediately, another example of the increasingly automatic ability to redirect attention without struggle.

Overall, the session was highly enjoyable and further strengthened the emerging impression that X5P1 possesses a distinctive operational style. Rather than the long, stable plateaus characteristic of X4P2, this cart appears to favour dynamic oscillations in both visual intensity and experiential depth. Whether this reflects the chemistry of a first pull, the particular cart, or simple coincidence remains unknown, but it has now appeared consistently enough across multiple sessions to justify recording as a working observation.

The possibility that future comparisons with X5P23C1 may help distinguish between first-pull and later-pull characteristics remains an intriguing avenue for future observation.
>>>
Much smoother, more flowing and coherent, all while maintaining the original info, and continued data.

Personal NOTE:
I use the same way of writing when I am doing more ''analysis'' type entries, or chapters for the book. I narrate what I want to say, have the AI search through all my past entries to compare, I then use it to do the back end research via published papers, indexed 3rd party trip reports, etc. All stuff that would be a time suck for me to do on my own. I then have the AI write my articles, based on what I said, after a clean-up, removal of obvious redundancies etc. I usually go theough 5 to 10 editing runs per article, post or chapter.
If anyone in interested in my canonized procedures, how I have set up the paramaters in GPT, voicing, how to use long term memory, indexing and project hierarchies, let me know. I'm happy to share if you have questions.

If, on the other hand, you feel offended by this or have any negative comments, I will kindly refer you to my customer service desk, of which all appointed employees are away for an indefinite amount of time! :)
PLUR
Q
 
Back
Top Bottom