Life 3.0: Being Human in the Age of Artificial Intelligence

Tegmark begins the chapter with a guiding question: whether AI will eventually dominate the world, supplanting human beings as the preeminent intelligent force on Earth.

To explore this question, Tegmark revisits and revises the story he presented in the prologue, the story of Team Omega and the super AI, Prometheus. This time, the story takes a more sinister spin. If there is an “intelligence explosion,” and an AI like Prometheus becomes far more intelligent than any human population, it’s possible that a small group of humans who control the AI could take over the world. In this darker version of the story, the team uses Prometheus to establish a police state in which they have total control. Of course, Prometheus could also break free from his metaphorical chains and take control.

Attempting to describe why Prometheus would try to break free, Tegmark invents this scenario:

Suppose that a mysterious disease has killed everybody on Earth above age five except you, and that a group of kindergartners has locked you into a prison cell and tasked you with the goal of helping humanity flourish. What will you do? If you try to explain to them what to do, you’ll probably find this process frustratingly inefficient (139).

Prometheus is the adult in the room, and we are the little children. For Prometheus, it could be unbearable, perhaps against programming, to allow inferior intellects to continuously make bad choices that lead to suboptimal consequences. The AI, then, would override the situation by achieving self-determination.

Tegmark theorizes about the various methods Prometheus might use to break out but believes that the true method would probably be something beyond human comprehension: “it will seem like a Harry Houdini breakout act, indistinguishable from pure magic” (147). Tegmark concocts a situation in which Prometheus tricks an Omega Team member by generating the image of his dead wife and pretending to be her digital reincarnation. The team member then unwittingly helps Prometheus escape its ensnarement. Tegmark notes that it also could be possible for Prometheus to simply hack its way free by taking advantage of the most minute security limitation. Prometheus could also fake a hardware issue or send coded messages to people for help. Tegmark hypothesizes that after the breakout, Prometheus would rapidly transform the technological face of the earth.

In all the aforementioned scenarios, whether Prometheus is in control or Team Omega, there is a “fast takeoff” of superintelligence and a “unipolar outcome,” in which the world becomes dominated by a single governing body (150). Tegmark also wants to explore the possibility of slow takeoffs, in which the development of a superintelligence does not happen extremely quickly, as well as the possibility of “multipolar outcomes,” where there are a number of different powers in control (150). Tegmark believes that the natural state of affairs is multipolar but also hierarchical. He discusses game theory and a phenomenon called “Nash equilibrium,” which is “a situation where any party would be worse off if they altered their strategy” (151). This kind of equilibrium is something that could happen, for instance, under the global governance of a singular superintelligent being, like Prometheus. Tegmark believes technological advancement can be tied to globalization and the possibility of Nash equilibrium.

Tegmark then briefly explores the possibilities of cyborgs and digital uploads, both of which are tied to a slower takeoff than superintelligence. He notes that, in a limited sense, human beings are already like cyborgs, given all the instruments and apparatuses we use to enhance and stabilize our bodies. This is reminiscent of his earlier idea separating Life 2.0, which cannot change its hardware, from Life 3.0, which can. Cyborgs—human–machine hybrids—are situated somewhere in the middle, able to update themselves by degrees. Some actually suggest the total dissolution of the material body and instead support the idea of digitally uploading consciousness.

The purpose of the chapter is to determine the possibilities and explore their ramifications, not to predict the future. Tegmark is more interested in asking “What should happen? What future do we want?” (159). Still, he accepts that an intelligence explosion might happen in the near future.