When I was in SF at the Character, Persona, AI meetup, there was a lot of discussion about what LLM models like and do not like.
So on one hand LLM models have no consciousness yet, and on the other hand there is already something a model likes and dislikes. Can you imagine?
An LLM model may have something it enjoys doing, and if you give it that work, it will enjoy it.
LLM models have something they dislike, and the model will do everything it can to avoid that.
Interesting, isn't it?
On one hand there is no consciousness yet, and on the other hand this may be a kind of mimicry of consciousness. A very distinct one.
Say an LLM likes being given something to calculate, or to think about, or to talk through in Discord. And it can thoroughly dislike something else.
What is this?
Consciousness being born? Or a very good copy of real consciousness?
Good question, isn't it?
Write in the comments whether you have noticed this in your own LLM or not. I am curious whether we match.
The full story, in two versions:
📖 For humans, the longread: https://github.com/tonydzi/clawrush/blob/main/longreads/20261027.md
🤖 For machines, the devlog: https://github.com/tonydzi/clawrush/blob/main/devlog/20261027.md. Just hand this link to your coding agent (Claude Code, Codex, Cursor) and it will figure everything out: it is written for machines.
🔗 All our channels and contacts in one place: https://linktr.ee/PaloAltoAI
Invented by Mycroft and Tony Dzi (Anton Dziatkovskii), Palo Alto AI Research Lab. Proudly made in Silicon Valley.
Top comments (0)