by James Wallace Harris, Thursday, June 29, 2017
I’ve been a science fiction fan my whole life. For sixty years I’ve waited for various science fictional concepts to come true. One of my favorites is intelligent robots. Around the time I discovered science fiction watching old movies on my family’s black and white TV scientists were inventing the concept of artificial intelligence. Back then, the 1950s, they had great hopes and made bold predictions. Over the years some of their predictions have come true, but not the technological singularity when machines become smarter than us. They could still become self-aware, but what if they don’t have to, what if they become much smarter than us even without sentience?
Yesterday I was reading about David Cope and his computer program Experiments in Musical Intelligence (EMI) in Homo Deus by Yuval Noah Harari. Harari described a challenge to Cope from Steve Larson, a professor of music. He proposed playing before an audience a real Johan Sebastian Bach piece, a piece composed by EMI imitating Bach, and a piece composed by himself. After the performance, they’d ask the audience to identify the composer of each. The audience thought the EMI piece was Bach, the Bach piece by Larson, and the Larson’s piece by EMI. You can read Harari’s “The Mozart in the Machine” for more of what he has to say, but I think it’s far more illustrative to listen to EMI.
This is rather beautiful – but is it art or creative? EMI is just a computer program that analyzes music styles and then imitates those styles. On one hand, it says our creative works have set patterns. Was Bach aware of those patterns, or was his composition a work of his unconscious? Obviously, EMI is an unconscious machine that composes.
In the 1950s when AI was new, scientists claimed if a computer could play chess it must have the special qualities of being human because playing chess is such a complex human activity. When Deep Blue beat Garry Kasparov in 1997 humans decided that chess playing wasn’t that special.
Here is a piece by EMI in the style of Vivaldi. Doesn’t it feel like EMI has captured something special?
I imagine, but I am not sure, that brilliant human composers could imitate other composers in the same way. Harari’s point is EMI composes music that moves human listeners emotionally. That somehow the computer program can capture the sublime. Of course, we like to assume our sublime experiences are the most complex and deepest of our lives. Isn’t EMI, maybe with the aid of deep learning, just figuring out how to push our buttons? How simple was it?
Homo Deus is an impressive book, but also disturbing. On one hand, it could be a handbook for a masterclass in science fiction writing. On the other hand, some could feel it’s like Biblical prophecy predicting the end of humanism. We live in a time after the Enlightenment where a large part of the world still accepts Old Testament thinking. So when Harari says liberal philosophy and humanism will be supplanted by techno-humanism it’s hard to believe. Won’t the world be 70% Old Testament thinkers, 20% humanists, and 10% techno-humanists?
What happens when we have true AI? What will the world be like with 90% unconscious machines, and 10% conscious? As Harari points out, humanism is based on the idea that all people are equal and they all deserve equal rights. But will biologically/genetically enhanced people feel that way? Will Human 1.0 accept Human 2.0? Will both of them accept AI 1.0? What will AI 1.0 think of Humans 1.0 and 2.0?
Corporations are backing robots over people. Capital is shifting to very few humans, and they want to eliminate all labor. Futurists talk of guaranteed minimum incomes, but capital doesn’t even want to pay for universal healthcare, so why would it support tax money going to completely support humans who can’t find work in a cyber economy?
Although I loved reading science fiction all my life, I’m not sure I’ll like actually living it. I thought my science fictional future would involve me traveling to Mars. Or owning a robot that did housework. But it looks like robots will colonize space, and take over all our jobs on Earth.
What are we suppose to do? Go to live in a virtual reality? Meditate and find our inner selves? Become artists? As Harari points out with EMI, robots will outdo us as artists too.
It will be fascinating to read science fiction stories read by writers studying Harari. If you belong to a species third down from the top how do you redefine existentialism or religion?
3 thoughts on “Is it Science Fiction Yet?”
To me, these musical pieces are more evidence that our sense of the transcendent is something that arises from our nature, not something inspired by some external force or entity. Recently, hardly a month goes by without some evidence that our animal cousins share some “unique” trait of ours, but will we ever know if they share that one?
Re SF’nal topics: it’s been a while since I noticed anyone talking about arcologies, although people in Japan, Germany, and Dubai have actually built similar structures to enclose amusement parks. Are domed cities due for an SF/real world comeback?
Nice thoughful entry, James
I am interested in similar topics of technology and AI, so I came across this post. I like your ideas and I think a lot of these questions will soon be very relevant for all people. But I personally think that I will like living in the new “science fiction” world! That is, if we (humans) execute it right. I think when AI comes, we just need not forget about humanity and being kind, setting it as the trend in the age of automatisation (and hopefully encouraging acceptance of all different kinds of humans/intelligent species instead of division). Thank you for an interesting read, made me think.