There has been a spate of recent books and media buzz warning of the rapid rise of artificial intelligence (AI). The fear is that this technology one day might not only surpass humans in intelligence but possibly take over the planet. Or — be able to thrive on a planet we’ve made too toxic for our own survival. But in Augmented Mind: AI, Humans and the Superhuman Revolution, author Alex Bates, an AI researcher and developer, is announcing the glass is more than half full. He seems to think we’ll both manage to control and harness AI as we collaborate with it to greatly improve the quality of all life, perhaps starting with strategies that will ensure our personal health, prosperity, and job satisfaction.
Augmented Mind begins with a survey of research into how the human brain functions. Then it moves on to describe the history of computer-based artificial intelligence theory and development. Today, the technology is advanced and continuing to evolve in myriad forms — from the processes behind Apple’s Siri to Google’s driverless cars to Bates’ Mtell, a system that can predict device failures and prescribe preventive repairs in complex manufacturing operations.
Bates is therefore not an objective observer here, but his professional credentials are impressive, having published in peer-reviewed journals while he was still an undergraduate and then contributing as lead inventor to three patents in the area of sensor networks and machine learning.
Perhaps the most intriguing (and potentially worrisome) possibility Bates describes is the entity he calls the centaur, after the Greek mythical being, a hybrid of man and horse. (The centaur will be familiar to sci-fi fans as the cyborg.) As Bates describes these superhuman entities, they will combine human intelligence, machine intelligence, and human intuition. Centaurs might not necessarily be individuals — they could well be expert teams of humans and computers. He speculates that, like human experts, centaurs are likely to be constructed as specialists. He envisions mega-genius centaurs in the fields of healthcare, personal assistance, chess and game theory, financial analysis and investment, engineering design, fine art, music, and (yikes!) writing, along with psychology and psychoanalysis (lest we have difficulty coping with change).
Computers are already assisting humans in all these areas, in ways that seem sophisticated and, well, intelligent. I didn’t find any of this particularly surprising, but I find the following statement stunning:
There may be problems so vast and so complex that it will be necessary to employ a variety of minds to solve them, including animal minds, machine minds, and centaurs.
Animal minds? Bates mentions this only in passing and does not elaborate. But it’s well known that biologists have had some rudimentary success communicating with primates and dolphins. Elephant communication is an active area of study in the field. These large-brain sentient species also exhibit signs of not only high intelligence but also self-awareness. For example, an elephant can recognize itself in a mirror, a classic test that the animal knows it is an individual.
As Bates reports and recent news articles confirm, Elon Musk’s researchers are developing Neuralink, a product that will permit connection of a tiny, wearable computer directly into the human brain, and with minimally invasive laser surgery.
So none of Bates’ speculations seem at all far-fetched. The main question for nontechnical people — as well as for public policy — is how to react to the changes.
Some observers believe that our increasing dependence on our smartphones — along with the societal changes brought about by the Internet — are just an early transformative stage in fusing humans with thinking machines. So, if the popularity and ubiquity of Facebook and Twitter are indications — perhaps we’re already baby centaurs?
Although Augmented Mind explains brain structure and modes of human thinking in depth, I didn’t find any scientific theorizing about the nature of consciousness itself. On this topic, Bates is more pragmatic than philosophical. In fact, he asserts that (perhaps avoiding the term consciousness) self-awareness can at times be a handicap to achieving a complex task:
For example, one does not want to be operated on by a neurosurgeon who is contemplating a change in career or distracted by a fight he had that morning with his son. AI lacks the self-consciousness that distracts the neurosurgeon…
To achieve successful social integration with AI systems, Bates believes those systems must be developed so as to be able explain the rationale for their decisions to humans. Here is the essential fear, I think — that some congress of centaurs will one day tell us to “push the button” and then refuse to explain why because we lack the ability to understand the answer!
Featured drawing by Dr. Johannes Sobotta [Public domain 1909]. Book cover and author headshot courtesy Alex Bates. Kindle ebook provided by NetGalley.
Gerald Everett Jones is the author of nine novels, including Clifford’s Spiral about a stroke survivor who tries to piece together fragments of his life, loves, and lusts. He is the host of the GetPublished! Radio show.
Be the first to comment