Here’s an odd thing to ponder if you’re tired of impeachment theater. Big tech continues to pursue the elusive dream (or nightmare, depending on who you ask) of achieving Artificial General Intelligence, or perhaps even “Strong AI” – an AI entity caple of possessing consciousness. To date, none of them have succeeded, at least that we know of. (Yes, some people believe the AI has already “woken up” but it’s hiding in the internet so we don’t discover it, but that’s a debate for another day.)
One possible problem, at least according to some tech experts, is that we’ve been trying to model Artificial Intelligence on the human brain. Why is that a problem? Because our brains are too complicated to replicate, at least for now. But there are other intelligent species with powerful brains hanging around. Why not model AI on one of those? One candidate being suggested is the octopus. (Boston Globe)
Many believe that mimicking the human brain is the optimal way to create artificial intelligence. But scientists are struggling to do this, due to the substantial intricacies of the human mind. Billye [the octopus] reminds us that there is a vast array of nonhuman life that is worthy of emulation.
Much of the excitement around state-of-the-art artificial intelligence research today is focused on deep learning, which utilizes layers of artificial neural networks to perform machine learning through a web of nodes that are modeled on interconnections between neurons in the vertebrate brain cortex. While this science holds incredible promise, given the enormous complexity of the human brain, it is also presenting formidable challenges, including that some of these AI systems are arriving at conclusions that cannot be explained by their designers.
The linked article is lengthy and rather dense, but definitely worth diving into if you are at all interested in this subject. The author, Flynn Coleman, may not be a computer scientist but she’s proposing some interesting ideas that are already being explored by experts in the field. In terms of why our brains are so difficult to recreate in a computer, she lists some of the aspects of our own brains that we don’t even understand yet. These include:
– We don’t know exactly how we make decisions
– We don’t have an accepted definition of what human intelligence is
– We don’t exactly know why we sleep or dream
– We don’t know how we process memories
– We don’t know what consciousness is
– We don’t have an equation to define what we call “common sense”
When you consider how much we don’t really understand about how our own brains work it’s no wonder we can’t teach our computers how to replicate them.
But if we’re looking for a different model, is the octopus really the way to go? They apparently have impressive brains to be sure. They are spread out, with much of the brain existing in the legs and different parts can work independently or in tandem with the rest. But since we can’t really communicate with them (yet) it’s probably going to be hard to reverse engineer their gray matter. Besides, some scientists are pretty well convinced that octopuses are actually aliens.
The octopus just seems frightening and, well… alien to me. Do they understand concepts like compassion or empathy? If we’re going to “wake up” the AI one of these days, I’d rather roll the dice and hope that the newly sentient technology at least has a chance of containing some compassion for us before it starts rolling out the terminator robots at the first automated car factory it takes over.
But I don’t want to be too hard on the octopuses. For a different and more intriguing look, here’s a video of an octopus dreaming. You won’t be sorry you clicked.