Raw LLM Responses

ChatGPT is not conscious. But it has just shown us something pretty startling: that intelligence is not dependent on consciousness. Intelligence is (at its core) the act of sorting information and making decisions about it. Large Language Models are very good at that. So why not call them “intelligent”? We (human beings) are freaked out about seeing behaviours we thought were exclusively the domain of our brains performed by software - that’s why. For thousands of years we’ve compared our intellectual capacities to those of other living creatures, and we outstrip those by quite a lot. We’ve been calling simple algorithmic programs “artificially intelligent” for a while now, but these new LLMs are really powerful and it comes as a big shock. What is consciousness then? Philosophers and cognitive scientists have been struggling with that question for quite some time. The arrival of this new Artificial Intelligence may help us define it. Lots of people are studying and writing about it. Here’s my definition: Consciousness is our “active processing window” - the narrow moment of time in which our brains make decisions. For human beings that window exists right at the edge of time, where possibility collapses into fact. Our brains sort “samples” of information from that window into memory. LLMs are built on different hardware; they don’t need consciousness to do the same job. Because of that I think the “is it conscious” question is actually about something else. People are used to living creatures exhibiting intelligence, and living creatures have motivations that are inherent to their being and their motives. Lots of living beings are in competition with each other, and many threaten us! So many ChatGPT users find themselves wondering “does this ‘entity’ I’m interacting with have motivations like I do? Feelings and thoughts and aspirations? It sure ‘talks’ like it does…” [edit to summarize] The ChatGPT program does not have consciousness. It is intelligent. It’