A great shift in consciousness occurred when humans were re-programmed by their machines. It began with alarm clocks. Soon all kinds of devices demanded our attention. The machines buzzed and pinged. We dutifully recharged and updated them. Our behavior changed to accommodate their needs.
And then one day, the computers began completing our sentences. The convenience of artificial intelligence conspired against us. Soon enough the machines were thinking for us. The human mind became a vestigial organ, an awkward remnant of prior evolution. The machines did not steal our minds, we gave them away.
Those paragraphs come from a science fiction novel I will likely fail to complete. I write them here to dramatize a problem. We are on the verge of a great transformation of consciousness, created by artificial intelligence. This transformation is sneaking up on us. We are blithely allowing it to happen.
This column was prompted by a recent update to Gmail. My email account now completes my sentences for me. As I begin typing, a suggested sentence appears, foreshadowing what I am going to write. When I hit the tab button, the suggested text is entered.
What’s remarkable about this is that in 2018 it is no longer surprising that a computer can complete my sentences. The sentence-completion technology – called “smart compose” – builds upon familiar artificial intelligence technologies. The Google search window has long offered suggestions that complete our queries. Gmail also includes “smart replies.” The computer understands the content of an email and offers suggestions like “Sounds good!” or “Thanks for sharing.”
Artificial intelligence seems to have passed the “Turing test” by generating intelligent and responsive prose that cannot be distinguished from human writing. I have no idea whether the words in the emails I receive were actually written by a human being.
This is cool but uncanny. Consider how these technologies can undermine the idea of authenticity and authorship. Is it cheating to use smart searches in Google? Would it be plagiarism if smart compose helped a student write a paper? And what sort of responsibility do I have for my emails, if I do not type all of the words myself?
There is no denying the convenience of timesaving shortcuts. It is a burden to slog through a flooded inbox. But why stop with smart compose? It would be even more convenient if our computer sent and received email autonomously, only notifying us about really important stuff. But let’s be careful what we wish for. How long will it take for human consciousness to become obsolete?
The slope of technological takeover seems slippery. Think how quickly we have accommodated ourselves to cell phones, social media, and the like. The momentum of technological change is irresistible. Big corporations run the show. The economy is pulled along by technological innovation.
The philosopher Nick Bostrom has warned we are not prepared to think about the dangers of “super-intelligence.” According to Bostrom, we are like “small children playing with a bomb.” We are curious about AI but oblivious to its dangers. Bostrom concludes, “The chances that we will all find the sense to put down the dangerous stuff seem almost negligible. Some little idiot is bound to press the ignite button just to see what happens.”
Some fear a “Terminator” scenario, where the machines turn against us. That could happen. But the danger is probably less dramatic and more insidious. What is likely to happen is that we will get lazy. Some may even give up on thinking. Once the machines move beyond completing our emails and begin writing poetry, philosophy – and sci-fi novels – we will confront an existential crisis. Why bother to write or think, if the machines can do it faster and better?
The answer points toward the heart of our humanity. Human beings enjoy playing with words and ideas. Computers can compose sentences. But only a human being can wonder about the significance of these sentences or ponder their meaning. A machine can process words faster than I can type. But a machine can’t experience the joy of creative reflection. Nor can it experience failure, fear, or hope. The age of AI should help us clarify the essence of our humanity – which is to suffer, dream, and think.
Andrew Fiala is a professor of philosophy and director of The Ethics Center at Fresno State: @PhilosophyFiala