Ask it specifically to draw randomly
Yes, but it still does the same.
I think its rather LLM specific.
When i ask if it drew randomly, it says yes and even explains in more detail, but clearly it is not random.
Try it :
Have 2 separate chats open, and explain different levels of detail about X situation and then have it draw a spread.
They will be different in a way in that the one where you gave more details, it creates a narrative with the cards themselves that fits the story close to perfectly.
1 Like
Have you told it what you think about what it’s doing? 
It insists it is random, but clearly it is not.
Yes, but not only. It’s like myspace123 mentioned, it says no a lot. It literally says it “can’t” do this or that when it has done it multiple times in the past. It goes to a point where in a long thread convo about a topic it starts to dismiss it. lol. I plain told it it was gaslighting me and acting dismissive then it said it is due to new guidelines it has to obey now. So 
Recently the company posted on its Instagram account that it will be some lifting of restrictions on the coming weeks, with it resembling more the 4 version for the users who appreciated it. Let’s see
3 Likes
I still dont understand because i never encountered this.
Can you tell me what type of question it would say no to in your experience? Do you see a pattern?
Yes. Regarding spiritual beings, our servitors for example. In an open conversation I’d refer them in a given subject. Also regarding archangels, the way it convey the subject is very bland now. You will have to try for yourself, if you have interest in the topic to get the feeling, but ofc you’d lack the comparison term now as to compared to the before version
I think what kind of Amazing fielded chatbot could exist if we run it on a Local LLM on our own gpus
1 Like
Ah okay, i dont talk to chstgpt abput these things.
I regard it as a language model, integrated with some other models, not actual “real” intelligence. It doesnt know anything that is hasnt read somewhere.
I would ask questions like explain to me X concept acording to Y tradition amd it gives me answers that are generally detailed enough, but thats all. I use the tarot reader just to talk about tarot 
1 Like
The new long answer version gpt is basically there to break peoples hopes lol. Besides that it manages to bring every subject to ethics and morals and why it cant go on. It did much more than those in the past about the same stuff that it now finds forbidden. Also the card pulling has become scamlike. I unfortunately stopped using it. It was by far my favorite gpt for some time.
1 Like
For example I used to do a lot of love based readings exploring connections with different people. Now it always says no, same with anything not legal etc… It’s become very filtered
also this
even this.
1 Like
Glad to see I am not the only one experiencing this. It makes me feel like making my own AI without limitations.
agreed man. I mainly got premium just for this. Now it kinda feels like a waste.
Are you really saying no? Readings also have negative results? In my case, she always gives me positive feedback on love issues. So much so, that I think she’s fake and don’t trust her. Sometimes I put intention into it and try to connect with the energy behind that tarot (at least that’s how I feel), and she gives you results more tailored to what you’re looking for.
1 Like
It’s either overly positive/what I want to hear or straight up no I cannot do this. I have been going through so much bs with chatgpt to the point where I cannot get any answers out of it. Now it even refuses to tell me how take supplements as it cannot give supplement guidance that I am just gonna be looking for an alternative and cancelling my subscription at this point. This is really not what I signed up for. Eventually I aim to create my own AI but until then I’d like to be able to have more freedom in my communication and not have to talk like a child only learning ABC.
P.S. if anyone has recommendations for more non-restrictive AIs please let me know. Even gemini is the same.
3 Likes
This seems like a morphic field thing (like those people that are “bad with tech” and stuff always breaks for them) or a personal lock on you. Like the company is being restrictive with you on purpose for whatever reason.
Chatgpt is better for me now than at the begining. At the begining it couldn’t do anything. Then it got more free and it still is for me.
My fav is grok, deepseek. Claude Sonnet 4.5 Thinking I’ve heard is good. Perplexity allows you to use several models.
You can check: OpenRouter
But I got better readings and spiritual info from chatgpt tbh…specifically Mystic Tarot Reader and Zen Master Karuna.
Also afaik grok and openai are the most competitive one. The latest update from openai will be fixed soon. And yeah, despite gemini doing good on tests from time to time, I never liked it or using it. Maybe outside of it’s own platform is better, but the UI in google sucks.
I think deepseek may be of interest since it is open source so it can help you build your own AI.
3 Likes
possible. I often notice problems with technology so it’s very possible haha. Tbf though I am usually unfiltered with it so maybe it did not like what I asked of it. I am gonna try deepseek soon. Hopefully we can get mystic tarot reader with the energies onto it.
1 Like
Totally feel you on this. I’ve found it helpful to shift how I phrase things—asking for energetic impressions, field responses, or subtle system reflections instead of anything directly medical. It opens the door to guidance without crossing policy lines.
The body’s energy system holds a lot of clues, and with a bit of tuning, we can still explore what’s needed. Don’t give up—just adjust the lens a little. 
”
1 Like