Sometimes, our words say more than we mean. Or, rather, we say more than we mean when the words we do not say are factored in. That is, what we imply matters more than what we actually say.
I largely agree with what François Chollet said last year as to why there will be no explosion of general artificial intelligence. But when he challenged the fear of an AI-driven “intelligence explosion,” he, perhaps unwittingly, said more than he meant.
Chollet, discussed here, is deep learning researcher at no less a firm than Google and Google is the proud owner of Deep Mind, whose AI program AlphaGo beat the world-ranked Go champion a couple of years ago.
Given that background, you’d expect Chollet to be, well, more effusive about explosive AI growth. He’s not; at least not when it comes to AI morphing into an all-encompassing, humanity-consuming super intelligence straight from sci-fi. His reasoning is straightforward:
Intelligence is not a knob we twist that correlates with accomplishment. Intelligence, if anything, is circumscribed by the culture and environment in which it lives. He observes that the supergeniuses we know of—those whose IQs (for whatever that measure is worth) leap past the rest of us into the 170s and beyond—have led rather banal lives. More often, humans who have made a difference, for better or worse, have had, by comparison, paltry IQs: “Feynman reported 126, James Watson, co-discoverer of DNA, 124 — which is exactly the same range as legions of mediocre scientists.” So-called “raw intelligence,” if we can even define it, is, at best, a piece in the puzzle. Below a certain level—one much lower than we might expect (70)— lack of innate intelligence inhibits effectiveness. But, beyond another level (130), its increase shows little impact on effectiveness.
What does make a difference then? Chollet presents a basketful of factors but two should give us pause. The first is that intelligence is part of our physical being, extending through our bodies. The second factor is the culture, or world, in which we live. I agree. But there’s more here than Chollet realizes. Consider his statement:
The brain has hardcoded conceptions of having a body with hands that can grab, a mouth that can suck, eyes mounted on a moving head that can be used to visually follow objects (the vestibulo-ocular reflex), and these preconceptions are required for human intelligence to start taking control of the human body. It has even been convincingly argued, for instance by Chomsky, that very high-level human cognitive features, such as our ability to develop language, are innate. François Chollet, “The impossibility of intelligence explosion” at Medium
Chollet is right to recognize that we, like all animals, come pre-wired. Young deer stand, leap, and run within hours of birth. Birds build nests without prior instruction. Squirrels bury and find nuts. We speak and juggle abstract thoughts. But basic chemistry does not create language; while speaking may require chemical bonding and signaling, language rests on something more. Vision is another “chicken and egg” problem: The best human eye in the world is worthless without a nervous system to transmit the signals and a mind to interpret them. Chollet recognizes that these abilities are “required for human intelligence” so that it can make use of a human body. What he neglects to ponder is whence did they come? How does intelligence arise? Chollet raised the question without answering it.
He, unintentionally, introduces the same tension when he discusses feral children— children raised, literally, by animals in the wild. Unless rescued young, these children never develop into full, social humans; they remain unable to use human language. His point is that we require culture to develop culture and use our intelligence. I think you can see the problem: If we need culture to function, then whence came the culture in which we have all been raised? Which is the chicken and which is the egg?
I strongly support Chollet’s core observations. The super-intelligent AI myth is little more than a replacement “god” for those uncomfortable with traditional theism. It is an article of faith. And, like all uncritically held articles of faith, it induces blindness: blindness both to the real problems AI can cause (when we cede unwarranted control to the machines) and to the stunning magnitude of the human mind.
Let’s ditch the myths and see the mind for all that it is, even if it leaves us pondering more difficult and unexpected questions.
Brendan Dixon is a Software Architect with experience designing, creating, and managing projects of all sizes. He first foray into Artificial Intelligence was in the 1980s when he built an Expert System to assist in the diagnosis of software problems at IBM. Though he’s spent the majority of his career on other types of software, he’s remained engaged and interested in the field.
Also by Brendan Dixon: What Does It Mean to Be Intelligent? (Evolution News and Science Today, 2016)
“Bob Dylan” is a mechanical mockingbird (Evolution News and Science Today, 2017)
See also: There is no universal moral machine Brendan Dixon’s view of MIT’s Moral Machine is featured.