3 Comments

Word Wooze will end us all.

Expand full comment

"... some organizing conscious intelligence..." Humans relentlessly anthropomorphize, betcha it's basic to how we think. It doesn't seem to be an impractical thing, I don't know whether you can even be a toolmaker/engineer without ascribing agency to the materials/machinery ... the stuff is a thing which is trying to do something, or at least it's supposed to do something. A "personality" seems a reasonable way of organizing some random set of interesting qualia/processes, potentially very abstract.

Expand full comment

an increasingly difficult thing to articulate is that as we advance our understanding of what consciousness and intelligence even is, we are finding smaller and smaller individual units capable of having memory, agency, and intelligence. This isn't an argument to say that LLMs are conscious beings (they aren't) but it is becoming increasingly untenable to assume that a brain is required in order to do the things we attribute only to conscious beings.

Cells alone can solve problems: https://www.scientificamerican.com/article/brains-are-not-required-when-it-comes-to-thinking-and-solving-problems-simple-cells-can-do-it/

Taking it further, one interpretation of the probabilistic motions of individual particles is that they are making decisions. Conway's Free Will Theorem and all that. I don't know, building consciousness entirely from matter that has no experience whatsoever of its own is becoming its own kind of implausibility. At some level, the universe itself probably has some sort of consciousness, if not agency. So it would only make sense that we would mistake this in LLMs as evidence of an individual agent making choices, even if it is very much not true.

Expand full comment