My morning routine includes waking up, making a cup of tea, and sitting on the couch in our little apartment and listening to my favorite news podcasts. Then i usually take Dot out for her morning constitutional to the dog park about a half a mile away where she rolls in the dirt and eventually poops in the corner. I clean up after her and then walk to the edge of the river where Dot likes to wallow. Then we sit on a bench by the river and I listen to the birds and ruminate on the news.
Then its back home to where I have some more tea and I begin to work on which ever project I’ve got going and get ready to take Jan to whatever appointment she has that day with whichever doctor or therapist she has scheduled. Today she had both her physical therapy, and her voice therapy. Later in the week she will have her neuro-feedback and then sometimes an actual doctors visit. It’s pretty simple and usually pleasant when the traffic isn’t crazy.
Today I asked her how her voice therapist visit went and she said. “She told me that Parkinson’s disease wants to make everything inside of your head small. So my job is to make them big.” When I asked for a bit more she said. “It makes my movements small and jerky and my voice small and mumbling. My job is to make everything big and loud.”
Jan has been diagnosed with Parkinson’s for about twenty years, though looking at pictures it seems pretty clear that her brain was slowing down on the production of Dopamine for longer than that: photos show flat affect, a slackness in her face and a rigidity in one of her arms and not swinging her arms when she walked. Anyway, she has had this condition a good long while.
My news podcasts are full of stories about Artificial Intelligence. There is a lot of lose talk about how we are close to creating a machine that will think in a way that is bigger and louder than a human brain. Lately I have been ruminating about the brain.
Of course the Spector out there is that we are about to build a machine that in a sense will be an all powerful psychopath without feelings or a moral center which will be able to hack into any man made system and wreak havoc. (as if human’s never had the ability to wreak havoc without the help of a psychotic robot before) The questions arises, are these worries really founded, and it seems that in the future it is possible, but still I’m wondering if we are not projecting our dark worries onto complex machines, that don’t really think, but merely calculate on a huge scale.
Two things bother me about this. One: the idea that we can build an all new powerful brain when we still don’t understand our own home grown brains sitting on top of our necks. (let alone fix the ones that are broken) And two: our naivety at thinking that we can really built a brain that allows the machine to be if not sentient then have a will to act.
It probably is a bad Idea to keep pursuing building these powerful calculating machines, no matter how we think of them, not because we would be building a monster, but because we have monster’s inside of ourselves that will find representations inside these machines. AI computers are already biased against women and people of color because the people who build them have those same qualities whether they admit them or not. There is also a class bias built into our fears of these machines. Robots have been replacing workers for dozens of years now. And the freak out rate has been much lower than now. Because now white collar jobs are at risk of being replaced. Because machines can do the parlor trick of seeming to be creative when really they are just massive calculators, fast learners of patterns and can skip two steps ahead of a human brain when trying to predict old patterns of expression.
Clearly, I’m out of my depth here. The real scholars in this field are not engineers, by philosophers and psychologists. The truth is that soon enough we will be able to wire up a machine nearly as complicated as the neural network of a human brain. But we are still a long way away from building a machine that can grow its own brain as we do. For I believe that to be truly sentient you need a body that grows from a seed into a flesh and blood structure that feels pain and forges breeding relationships. We are a long way from being able to do that.
Just as we are a long way from building a machine that can leave work and sit in the sun and say, “Man I’m exhausted. I hope my partner is in a good mood when I get home.” A machine may say those words unbidden from a huge script of possible things to say but they cant feel it and create the expression from experience. You need a body to be sentiant.
Watching my baby grandson grow only confirms this. He struggles with language and understanding. You can ask him a question and you can see him processing, processing. “Arthur will you bring me the white bowl,” and after hard work, he turns and brings it to me. I don’t know how much of the request he understands. Does he know, “White” or “Bowl” I don’t know but he clearly recognizes that he lives in a physical context: there is me, and him, and that thing he was playing with in the sandbox. He lives in a physical reality and he knows through growing his own brain… which is different than transferring a massive dump of data. Which computers will always be better at than we are.
The danger with machines is that human’s are lazy and we’ve always built things to carry our load: wheels, shovels, pulleys, bombs but the danger of building machines that manipulate language is that we are filled with prejudgements, and contradictions. Machines will be able to destroy us only if we build them to do just that, and it seems that we should be able to place checks into our inventions to prevent our own destruction… but it will take an extraordinary amount of self awareness. Which of course is the problem and takes the issue out of engineering and into the world of humanistic contemplation. Coming all the way back to the question as old as the pyramids: Just because we can build something, does that mean we should?
Here is a short poem I wrote for a dance troop in Sitka, They were experimenting with the techniques of the great dancer/choreographer Merce Cunningham who would ask his dancers to improvise on a single scrap of writing. I wrote this for them and they made a beautiful dance.
