[labrador]

I had Claude Sonnet 4.5 and Gemini 3 do a Tarot reading for me given what their memory systems know about me and was underwhelmed but I that just may be Tarot, which I'm not familiar with.

More interesting was when I asked them to tell me which character I most resemble in the folowing: The Wire, The Sopranos, Better Call Saul/Breaking Bad and the game of Chess.

Both said I was Freeman in The Wire, Hesch in The Sopranos, Mike in Breaking Bad and a Knight in chess. I asked them to explain their reasoning, which was enlightening from a self-knowledge perspective.

[jackzhuo]

That 'underwhelmed' feeling is exactly what I was trying to avoid!

It makes sense that the character matching worked well—AI is incredible at pattern matching your past data/context. But Divination (Tarot/I Ching) requires Synchronicity and a sense of 'randomness' that feels earned.

When an LLM generates a reading, it's just predicting the next likely token, which flattens the magic. The manual ritual (shuffling cards or clicking stalks) restores that 'weight' that AI removes.

[labrador]

I like your take on this. It makes sense.

[manfromchina1]

They would likely pick the cards according to what your text looks like at the moment. You should at least ask them to use a random function. And even then they might keep picking according to the mood/content of the text.

[labrador]

Your statement would be accurate if Claude and Gemini had no memory system and each chat were fresh. I've been talking to Claude for two years so it knows quite a bit about me. Gemini less since they were late to add memory. Since your handle is "manfromchina1" perhaps Chinese AI don't have memory systems?