Andrew Witherspoon / Axios
What role will the human race play in an age when artificial and superhuman intelligence roam the universe? MIT cosmologist Max Tegmark argues that we are thinking about the question all wrong.
To start out, the Stockholm-born Tegmark, a co-founder of the star-studded Future of Life Institute, refuses to be limited by a conventional understanding of life. Rather than springing from metabolizing cells, life in a post-biological world, he says, will be "a self-replicating information-processing system whose information (software) determines both its behavior and the blueprints for its hardware."
This of course describes DNA—the basis for all known life forms. But it also makes room to see artificial intelligence itself as a living thing.
It's a question of consciousness: Life is one thing, but sentience — the state of being aware and thoughtful — is another. Tegmark has pondered the distinction before. Now in Life 3.0, his latest book, he writes that unless we understand consciousness, and whether we can artificially recreate it, we will not reach what we think of as AI. And we cannot know what our ultimate role we be.
Tegmark asks: "What's the point of uploading yourself into a robot that talks and acts like you if it's a mere unconscious zombie, by which I mean that being the uploaded you doesn't feel like anything? Isn't this equivalent to committing suicide from your subjective point of view, even though your friends may not realize that your subjective experience has died?"
His thesis: Life's first iteration was organisms whose DNA evolved over time to help them thrive—bacteria have learned to be resistant to antibiotic medecines, for instance, over the course of several generations. Humans today represent Life 2.0: our bodies (or hardware) not only evolve, and we can learn things like language (our software). Life 3.0 will be when humans and the intelligent machines we create dramatically improve our physical bodies, one generation to the next, bringing, among other traits, immortality.
Tegmark argues that we are already seeing life inch toward being "finally fully free from its evolutionary shackles"—we can already upgrade our "hardware" in the form of pacemakers, artificial knees and teeth. The laws of physics allow for Life 3.0, and the history of our universe suggests that such superintelligence will in fact be born.
Here's why:
- Current technology can replicate the power of the human brain, measured in floating point operations per second (FLOPS), at the cost of about $1,000. "There's no longer a strong argument that we lack enough hardware firepower or that it will be too expensive" to create an artificial brain as powerful as the typical human's.
- Embedded in the laws of physics is a drive both toward efficiency and the creation of ever more complex life forms. Now these complex life forms, namely humans, are creating complexity of their own. "Not only [does the human race] now contain more matter than all other mammals except cows, but the matter in our machines, roads, buildings and other engineering projects appears on track to soon overtake all living matter on Earth," Tegmark writes.
The potential of life as information: If life is merely an information processing system that can replicate, the potential for life to populate the known universe is unfathomable. The human brain developed to conserve precious calories and "be self-assembling, self-repairing and self-reproducing," but human-created life (and the life that our creations create) will not be limited by those requirements.
- Computers can perform the same information processes that the brain does, which means that life can take on forms much better suited to survive anywhere in the universe, in nearly infinite forms.
- Humans have come nowhere near unlocking the energy potential of the universe:
"Einstein taught us that if we could convert mass to energy with 100% efficiency, then an amount of mass "m" would give us an amount of energy "E" given by his famous formula E = mc2, where c is the speed of light. This means that since c is huge, a small amount of mass can produce a humongous amount of energy. If we had an abundant supply of antimatter . . . then a 100% efficient power plant would be easy to make: simply pouring a teaspoonful of anti-water into regular water would unleash the energy equivalent to 200,000 tons of TNT, the yield of a typical hydrogen bomb—enough to power the world's entire energy needs for about seven minutes. In contrast, our most common ways of generating energy today are woefully inefficient. ... Digesting a candy bar is merely 0.00000001% efficient, in the sense that it releases a mere ten-trillionth of the energy mc2 that it contains. If your stomach were even 0.001% efficient, then you'd only need to eat a single meal for the rest of your life."