There is one (of many) nigh inevitable paths a blog like this was destined to follow: I think one of those is leading towards the philosophical road to trying to wrap our minds around consciousness.

There are huge social and economic discussions to be had around AI, of course, but by far the more enduring one is bigger than will AI take our jobs—but rather asking, will AI take our place?

Hmm. Existential dread much?

Don’t mistake me: I am almost certainly not at all qualified to discuss this with any authority. That’s not going to stop me, of course, from asking a bunch of questions and trying to get my head around it. But I though it worth mentioning that my formal experience in this field was not much more than a few undergrad courses in philosophy and a lot of personal reading. I am a curious dilettante, but one with a small platform. I have spent a lot of time, in other words, thinking about thought, and now writing about it.

The source of my curiosity is hard to pin down: science fiction likely played a huge role, and I’ll swing my way back around to that in some future post, but I’m going to start in the middle and get right into the meat.

See, a year or two ago I had picked up a book, started reading it, and had only made it about three chapters in before I—well, I probably got distracted reading RPG-Lit or something—and it found it’s way to my bookshelf. That book caught my eye the other day, bookmark still protruding from where I’d abandoned it, and seeing it again triggered a thought: the author, I’d wager, probably has something interesting to say about artificial intelligence despite the fact that his thesis was written in the 1980s. It might be worth trying to trudge through one more time.

Roger Penrose released The Emperor’s New Mind^[1] in 1989 and essentially asked the question: Is there something going on in our brains or elsewhere when a human understands a thing—a phenomenon that cannot happen through computation alone? In other words, are our brains meat computers—or something more special? 

Penrose spends a lot of time digging into mathematics and physics, trying to give the reader a sense of the intuitive beauty of computational studies and to illicit a notion that something deeper is going on when we think about math than raw calculation. (This is where I got lost in my first read through.) He goes on and then seems to suggest that something ineffable is going on in our meat brains that is not readily described by computational processes—non-computational physics, bolstered by quantum effects maybe—and thus, perhaps, can’t necessarily be simulated in a computer.

In 1989 this was raw hypothesis. Average home computers were barely past the computational muscle of a Super Nintendo console. AI was theoretical research and the stuff of mysterious research. And science fiction did a lot of hand waving around artificial consciousness supposing that anything from clockwork mechanisms or a few microchips could do the job.  Nearly forty years later and now we house AI models in hulking power-hungry data centres that can be seen from space, models that require hundreds of gigabytes of memory to hold themselves resident and active, and when set loose can debate these very topics through a web browser on our phone.

…and so its tough to not see that we can create a pretty good simulation of many of the things that Penrose had argued were unique to those meat brains of ours.  

That doesn’t mean consciousness follows, but it opens the floor to asking some new and interesting questions—paths worth exploring on a blog like this-—about our own evaluations of what it even means to be conscious, and even to be human. Poets, indeed.