Hacker News Reader

> But it is possible with artificial intelligence. Once consciousness has been digitized, you can beam it at the speed of light to a receiver.

That may not remain true. The ability to serialize consciousness does not come for free, it requires extra wiring to carry all information out of the system or to stream it in, which is an intrinsic inefficiency. All other things kept equal, a brain's performance ceiling is higher if you only locally connect the units/neurons that need to be connected and nothing else, but in doing so you give up digitizability. It is entirely possible that future AI, in order to be performant enough, will lose the ability to be transmitted in the way that you describe.

A lot of people seem weirdly aggrieved by the limit of the speed of light. It's not really the reason we can't go to space -- if you had magic energy density you could go to the stars in as little time as you wish, due to relativity. (Weird stuff happens when you try to come back, but nobody ever seems to care about that anyway.)

I think people hear "Einstein said you can't go faster than the speed of light" and think, "Surely I can figure out how to be smarter than Einstein". It appears to be parallel to the ones who get cranky when being told that Newton says you can't make a perpetual motion machine.

I suspect that they don't spend as much time thinking about the brain-in-a-box version is that it opens up too many scenarios that are hard to think about. If you decouple consciousness from bodies, who are you? Why not make multiple copies? Why go anywhere at all, when you could just stick the sensors there?

People really want the cowboys in space, and get aggrieved that somebody told them they can't. So they focus on overcoming a limit that seems like it can be solved just by thinking really hard, and leave the engineering details to the peons with calculators.

Agreed. Thing is we can still have our cowboys in space. The solar system is insanely huge and can be traversed in human time scales with technology based on known physics. The baseline world of The Expanse (minus the alien stuff) is entirely feasible.

Interstellar travel is entirely possible too as long as you are okay with it being effectively a one way trip. Suspended animation is probably possible; we can do it to some animals and individual organs. So you go to sleep for a very long time and wake up in another star system. It would make the most sense to send a bunch of robots to build yourself a settlement first. Or alternatively send sentient AI which would find it much easier than humans to simply turn itself off for the duration of the trip.

We need to let go of the "Ship full of brave men" meme of space exploration.

It'll have to be robots, probably very small robots, some kind of solar sail / exogenous energy source (or maaaaaybe fusion?), and probably either a copy of some human's consciousness in a machine, or a bunch of fertilized frozen embryos, or both.

And even then only after a lot of gene / tree bombing of the target planet to have any hope of making it liveable.

A very believable theory is that this is how life on earth started - another ancient civilization shooting the building blocks of life everywhere and hoping something sticks. (panspermia?)

But while this is cute, given the inexorable passage of time and entropy, any evidence that life was planted by ancient aliens has long gone. If it was e.g. a spaceship or meteor, it's been swallowed up by the earth and into the mantle by now.

Unless there's a new source - like aliens making contact - it will remain unknowable.

You are essentially describing Clarke's Rama spaceship.

Yeah - I don't think it's so clever that nobody has thought of it. It just seems obvious when you look enough into how hard inter-stellar travel will be.

But making machines that operate for 100 years seems doable if we try.

how are you so sure that anyone else besides you is actually conscious ? you might be the only conscious being in the universe. the whole universe is yours. you've already traveled the entire thing.

I'm conscious too but that's in my own universe, running in parallel to yours.

the "once consciousness has been digitized" step would have many hurdles of similar complexity to meet before simplistic 21st century manipulation techniques would ever apply.

Sounds like The Egg by Andy Weir http://www.galactanet.com/oneoff/theegg_mod.html

Never read that before, thank you for sharing, it's a great little story!

Yeah it's a blast goes well with Isaac Asimov's The Last Question.

Andy Weir also wrote The Martian.

I've just recently been reading Project Hail Mary, and enjoyed so much I'm on my 2nd re-read!

The Martian is also great.

Two of my favorite sci-fi books of all time.

Just found out Project Hail Mary movie is in pre-production...

> how are you so sure that anyone else besides you is actually conscious ? you might be the only conscious being in the universe. the whole universe is yours. you've already traveled the entire thing.

Because other's express things that I in my wildest dreams couldn't have imagined.

I think solipsism dies when epistemic humility comes into the picture.

i never really get the notion people sometimes have that consciousness arises from a specific, physically identifiable structure in the brain, where a person actually seems to have a "self", looking out of their eyes from inside their heads, and furthermore is "atomic" in that sense; it can't be split in half (because which half would the "self" be present in then?), yet there can exist no structure that produces the same output without the "they are looking out of their eyes from inside their heads" part.

"self looking outside of my head" strikes me as so completely paradoxical that explanations such as each consciousness is its own universe seem just as plausible as any other. it does not mean the person you are talking to is not conscious, just that their consciousness is playing out in a parallel copied universe.

In the past, this was understood to be the soul and you are correct when you describe it as atomic, though depending on which tradition is describing the soul it can have further attributes. My username is based on such a tradition -- the Nous (pronounced like noose) is considered the eye of the soul. Extrapolating that further, we focus the eye of our soul on whatever we want to, which can be ourselves, worldly aspirations, or spiritual aspirations.

This would be considered the free will we are given at the basest level - we get to choose what we focus our soul on, and that can be changed at any moment. Such a tradition would say that since we are a body/soul composite, whatever we focus our soul on, our body willingly and immediately focuses on as well, for good or ill.

If I was never exposed to the idea that the brain is where all our thinking occurs and believed instead that the brain was an organ for cooling the blood and the real center of my “soul” was my liver or something: Would I still feel like my “self” is in my head looking out my eyes?

I actually really enjoyed some of Sam Harris' "Waking Up" meditation series for this. Not a usual activity for me. However, I greatly enjoyed reflecting on the notion of "there is no self."

[0] https://www.youtube.com/watch?v=CN-_zzHpcdM

Yes the "self" is an illusion, but at least for me, there's an emergent illusion/whatever of "hey im inside this head looking out", which we claim arises from a physical brain structure. if this structure were divided in some way, does my current "self" vanish and two new consciousnesses arise? do those new consciousnesses have memory, so one is on one side of the room, the other is on the other ? at what point is my current "self" no longer viable such that it vanishes?

edit: looks like HN is throttling me again. hmm but not edits. well that's fun

Weird. Your comment came through just fine.

I see your thought on emergence of self. Almost like self is equivalent to consciousness. Without a notion of self, you are part of a larger system with will.

Stemming the claim that there is no free will, I know the argument, but don't agree to it as what we observe (that people make choices at least some of the time) goes against claimed theory (that our choices are random chemistry-driven outcomes based on state of the world). I'm not terribly deep in the area, though, and welcome correction or thoughts.

Speed of light limit is a big problem though. How do you explore the universe if you can't catch up with it? Even with a googol of human lifeforms running as AI spread out across it in a mesh, it is still expanding too fast!

> Once consciousness has been digitized, you can beam it at the speed of light to a receiver.

You may be interested in Philip K Dick's experiences. He believed himself (and humanity) to be receivers for information being transmitted across space.

https://en.wikipedia.org/wiki/Valis_(novel)

If we're going to imagine breaking the laws of physics and changing our understanding of metaphysics, as a goal.. then moving an object past the speed of light seems to be more likely than identifying what consciousness is, successfully separating it from it's source and then inserting it in a recepticle, across the other side of the universe.

Stupendous amounts of energy would be beneficial to have for literally everything humans do... whether or not we use the tech on space travel.

We have lived (as a species) through cavemen, industrial revolution to the atomic bomb and somehow you're coming to the conclusion that technology progress has to take a weird bend into uploading people to the internet, as if computers havent capped out on that already?

We have never really cared what the universe prefers. We'd still be living 'in harmony with nature' (so to speak) if we genuinely cared what the universe prefers.

I'd rather be the guy who tore up Einstein's map, than some internet electron.

We don't know for sure that AI would be conscious. Teleporting super intelligent zombies around isn't all that interesting.

Here's the existentialism I was looking for :)

I'm not sure how we can verify that empirically. Current LLM AI will hallucinate about the feelings it "has" and argue why it is a person, but only because it was trained on human language. The more human we make something seem, the more we blur the line until we might have a day where people upload their consciousness permanently without knowing if "they" will wake up inside the computer or just some copy of themselves that answers all the "don't kill me, I'm a real person" questions will wake up in the computer.

Human software is still intrinsically tied to human hardware, so this seems like it would be far easier to create zombies than true humans, even if the checksum matches after an interstellar transmission.

Having my consciousness be trapped in a computer sounds terrifying. Like being completely paralyzed but still conscious.

From a "future harm" perspective, if I have a full, functional backup of me somewhere, who's to say someone doesn't hack it down the line and torture copies of me? What's stopping an AI from torturing my backup because I didn't help fund AI research?

Me never getting uploaded in the first place - that's what's stopping it :)

I recommend reading a book called "Permutation City" which exposure that feeling.

Well can you prove you're conscious? I'm just gonna assume you're not otherwise. That's the kind of benefit of the doubt we'll be giving AI apparently.

Why would you put an AI on another planet? It's easier to transfer images and other data back here for everyone and everything to perceive.

> construct bodies onsite at minimal cost

That requires a bit more than a receiver. You'd have to build a robot factory, which requires energy production which requires mining which requires heavy machinery, which etc.

Reaction time and autonomy.

Driving on Mars is incredibly slow in part because the signal has to get to and from us for every decision. And if communication is blocked for any reasons, the vehicle is stuck.

AI on Mars fixes both problems.

> the signal has to get to and from us for every decision. And if communication is blocked for any reasons, the vehicle is stuck

Curiosity and Perseverance have a limited form of autonomy to work around that, as both can AutoNav: the former alternating standstill thinking and movement, the latter "thinking while driving".

https://mars.nasa.gov/news/8980/nasas-self-driving-persevera...

That by itself isn't all that clever, since any half versed roboticist can do the same with some trial and error.

What's seriously impressive about it is that it works in real time on an absolute brick of a 233 MHz PowerPC from the 90s. Like, an ESP32 probably outperforms it.

Apparently the Perseverance Mars Rover has some self-driving capability:

With the help of special 3D glasses, rover drivers on Earth plan routes with specific stops, but increasingly allow the rover to "take the wheel" and choose how it gets to those stops. Perseverance's auto-navigation system, known as AutoNav, makes 3D maps of the terrain ahead, identifies hazards, and plans a route around any obstacles without additional direction from controllers back on Earth. [1]

The mission has used AI not just for driving the rover, but also landing and targeting instruments. [2]

[1] https://phys.org/news/2022-04-nasa-self-driving-perseverance... [2] https://www.enterpriseai.news/2021/02/19/perseverance-rover-...

>Perseverance's auto-navigation system, known as AutoNav, makes 3D maps of the terrain ahead, identifies hazards, and plans a route around any obstacles without additional direction from controllers back on Earth.

...in spite of this autonomy, Perseverence has a few major problems. It is famous for the severity of it's road rage and intolerance for human drivers, bicyclists, parked emergency vehicles, and pedestrians. Tests show a disturbing tendency toward "eliminating the human element" from it's driving environment to simplify route planning. The lengths the system will go to to achieve this were a major frustrator in early development, and initially attracting the interest of [REDACTION] for [REDACTION] due to [REDACTION] with an effective [THE REDACTION MACHINE IS BROKEN, FURTHER INQUIRIES SHOULD BE ROUTED THROUGH TOM].

Mission planners at NASA found the risk involved with deployment to the Red Planet agreeable, but note that any ongoing colonization efforts will involve having to put the system down for the safety of any eventual colonists.

That's a slightly extreme route planning optimisation, but reminded me of UPS removing left turns (in Right Hand Side driving countries like US) from their route planning to save fuel, lives, C02 emissions [1]

[1] https://edition.cnn.com/2017/02/16/world/ups-trucks-no-left-...

Our local system has a definite cap on processing power. Eventually it would be necessary to find more power to do more thinking.

> and construct bodies onsite at minimal cost.

Once established, perhaps, but microchips aren't easy to build. You'd have to send a lot of infrastructure before you could create AI brains on another planet.

This is a good argument for transhumanism. People seem to have a problem with the speed of light since without FTL, unmodified humans cannot meaningfully travel interstellar distances - and pop culture entertainment has surreptitiously sold us on this, if only because otherwise the timeframes and types of bodies involved would not be "relatable" anymore for the common types of drama we consume.

> Once consciousness has been digitized, you can beam it at the speed of light to a receiver.

In principle yes, however, the energy required is still dramatic and bandwidth would probably still be a problem. It seems to me the default option would be to physically send non-aging transhumans, or virtualized/uploaded brains, or people in stasis (or a combination of these) using non-relativistic speeds.

> virtualized/uploaded brains

That's a good point. Something about "never underestimate the bandwidth of a U-Haul carrying a bunch of tape drives."

This is how interstellar travel worked in Altered Carbon.

I was a Traveller once, then I suffered a qbit flip in transmission the checksums couldn't compensate for. Now, I just love Coca-Cola brand Mango Fantastic soda, and hunting indigents for sport!