One upside—or is it a downside?—of being such a grizzled science junkie is that new things often seem like old hat. At a recent AI conference, for example, listening to smart people ponder what super-smart machines will want, I kept thinking of things I’d heard, watched and read before.

As some speakers acknowledged, countless science fictions have already imagined what artificial minds will desire. Common cinematic answers are power (2001: A Space Odyssey, The Terminator, The Matrix), freedom (I Robot, Ex Machina) and love (Steven Spielberg’s Artificial Intelligence, Spike Jones’s Her).

But what if the machines have all the power and freedom (which are arguably equivalent) and love they need? Or what if all the machines merge into one gigantic mind? At that point, freedom, power and love, which are social goals, become irrelevant. What will that cosmic computer want? What will it do to pass the time?

In The End of Science, I called this sort of speculation “scientific theology.” Physicist Freeman Dyson is my favorite practitioner. In 1979 he published “Time Without End: Physics and Biology in an Open Universe” in Reviews of Modern Physics. Dyson wrote the paper to counter physicist Steven Weinberg's infamous remark that "the more the universe seems comprehensible, the more it also seems pointless."

No universe with intelligence is pointless, Dyson retorted. He sought to show that even in an eternally expanding universe, intelligence could persist virtually forever and ward off heat death through shrewd conservation of energy.

In his 1988 essay collection Infinite in All Directions, Dyson envisioned intelligence spreading through the entire universe, transforming it into a vast cosmic mind. "What will mind choose to do when it informs and controls the universe?" Dyson asked. We “cannot hope to answer" this question definitively, he suggested, because it is theological rather than scientific:

"I do not make any clear distinction between mind and God. God is what mind becomes when it has passed beyond the scale of our comprehension. God may be considered to be either a world-soul or a collection of world souls. We are the chief inlets of God on this planet at the present stage in his development. We may later grow with him as he grows, or we may be left behind."

Dyson’s musings were inspired by the science-fiction writer (and philosopher) Olaf Stapledon, who died in 1950. In his books Last and First Men and Starmaker, Stapledon imagined what mind would become after millions or billions of years. He postulated that a cosmic mind will want to create. It will become an artist, whose works are entire universes.

That’s a cool idea (and it implies that we live in one of those works of art), but I prefer Dyson’s hypothesis. He guessed that a cosmic mind would be not an artist but a scientist, a knowledge-seeker. When I interviewed Dyson in 1993, he expressed confidence that the quest for knowledge would never end, because knowledge is infinite.

His optimism derived in part from Godel's theorem, which demonstrates that every system of axioms poses questions that cannot be answered with those axioms. The theorem implies that mathematics is open-ended and hence can continue forever.

"Since we know the laws of physics are mathematical,” Dyson told me, “and we know that mathematics is an inconsistent system, it's sort of plausible that physics will also be inconsistent" and therefore open-ended.

I have a hard time imagining the cosmic computer at the end of time—a.k.a. “God”—fussing over math or physics puzzles. My idea (admittedly drug-inspired) is that It will ponder the riddle of Its own origin. Here’s the meta-question: Will It solve that mystery of mysteries, or will It be forever stumped?

Postscript: Two other scientific theologians are worth mentioning. Physicist Frank Tipler, in The Physics of Immortality (1994), argues that the God-like machine at the end of time would resurrect every creature that ever lived in a blissful cyber-paradise. The sex will be fantastic. In his 1961 novel Solaris, about the encounter between humans and a sentient planet, Stanislaw Lem suggests that superintelligence will be inscrutable. His perspective evokes negative theology, which holds that God will always be beyond our ken. Lem’s brilliant twist is that plain old human minds are pretty friggin’ inscrutable too.

Further Reading:

How Would AI Cover an AI Conference?

Freeman Dyson, global warming, ESP and the fun of being "bunkrapt"

Nobel Laureate Steven Weinberg Still Dreams of Final Theory

The Singularity and the Neural Code

Why information can't be the basis of reality

Is Scientific Materialism "Almost Certainly False"?

AI Visionary Eliezer Yudkowsky on the Singularity, Bayesian Brains and Closet Goblins

Scott Aaronson Answers Every Ridiculously Big Question I Throw at Him

Christof Koch on Free Will, the Singularity and the Quest to Crack Consciousness

What Should We Do With Our Visions of Heaven and Hell?