Verified by 
Artificial Intelligence
The Visitor From Planet X
Instead of sending messages to other stars, let's send computer programs
Posted July 20, 2012
When I was a kid I read the Tom Swift book The Visitor from Planet X (fulltext on Gutenberg) and even then I thought it was ridiculous. Tom Swift makes freakin’ extraterrestrial contact but then spends most of his time combating the evil Brungarians who are trying to abduct the Visitor.
Here’s how it goes: Tom gets a message from Planet X notifying him that a visitor made of “pure energy” is on the way. Tom’s job is to make a mechanical body for it to inhabit. This he does, and for some reason it looks like a Christmas tree. The energy duly arrives in the body and they celebrate, as you can see in the illustration.
Illustration from "Tom Swift and the Visitor From Planet X."
(I find this an exceptionally weird drawing, by the way. It looks like they’ve formed a cult.)
Great buildup, but after that the Visitor is quite the disappointment. All it does is move around the lab crashing into things, and emits terse sentences like DANGER NEAR. BRUNGARIAN SCIENTISTS READY TO DESTROY ME.
That’s it. No conversations about alien biology, technology, ideas, art, science, or ways of life. Tom never asks the alien where it came from. Or what it thinks of Earth. Even though I was 12 years old, I felt gypped. What a failure of imagination. I think it shaped my adolescence, actually: it gave me the belief that the grownups never tell you the really good stuff.
Admittedly, the Visitor might have had its priorities straight: survival first. It is impressively clear on its political allegiances:
DO NOT WORRY, MY FRIENDS. I WILL NOT RESPOND TO ANY ATTEMPTS BY BRUNGARIAN SCIENTISTS TO COMMUNICATE WITH ME. MY PLANET IS WELL AWARE OF THEIR DANGEROUS AIMS. HAVING CONQUERED YOUR WORLD, THEY WOULD NEXT INVADE SPACE.
This is sort of like an American taking sides between the House of Lancaster and the House of York during the Wars of the Roses in the late 1400s.
But now that I’m working on a proposal for a book about communication with alien intelligence, I’ve realized that The Visitor from Planet X actually had the seed of a good idea. Physical travel between stars is offensively difficult. Radio communication takes offensively long. But what if the program for an AI could be sent between stars? Then we might be able to send (and get) real “visitors from Planet X.”
This is the idea behind CosmicOS (“cosmic operating system”), in which we send not merely messages but programs to other civilizations. The programs will be able to interact with their curious alien inquisitors, and would—in theory—supplement the information that could be provided in a message. They might offer a simulation of some part of our ecosystem, for example. According to the CosmicOS website, an alien user “is free to play around with the simulated world and understand its logic through experimentation.”
Naturally there’s a lot of assumptions being made here. The aliens will have computers; the aliens will understand that the message is a computer program; the aliens will be able to translate it into code that actually runs on their computers, or create an emulator that runs the code as is. If you can get past all that—voila!
But why stop at running dinky little simulations? Why not send the code and state information of an AI? Now that would be a real visitor, able to converse with its hosts and share details of Earthly life and thought, and return with its new knowledge. For it the trip would be instantaneous both ways. Decades or centuries would pass on Earth, of course, but it’d still be faster than physical spacecraft and more satisfying than just exchanging greeting cards.
This is dependent on actually having AIs, of course. Ray Kurzweil (and a lot of other people) blithely assume that self-aware minds could exist in a computational substrate. I’m not so sure. Roger Penrose thinks that consciousness depends on quantum-level events inside neurons that couldn’t be modeled in classical computers. A lot of people disagree, but Penrose is smarter than nearly everyone who disagrees with him, so personally I wouldn’t count him out.
But let’s say that eventually self-conscious AIs do come into existence, by whatever means. And let’s also assume that their software, and their state data, can be fully characterized in a digital stream of 1s and 0s. (This is exactly what Penrose disputes, of course, but let’s go on.) And let’s say we preface the AI with an introductory message explaining how to instantiate it on any reasonably powerful local machine.
If I was an AI, I would find this insanely exciting. A whole world of aliens to explore! And I'd be there in the blink of an eye, subjectively speaking. The trip back would be just as fast, and my AI friends would probably still be around to hear the story.
That's all a lot of ifs. But that could be how we will someday create a real Visitor to Planet X.
Liked this posting? Read my previous one, Will Extraterrestrials Understand a Message We Send?
