Fascinating paper, I'm going to have to spend some more time with it later today. I did love this quote though which captures the ethos of what they are proposing:
>: Our asteroid restructuring approach can be compared to the effort of early pioneers. A wagon heading west in America in the 1800s could
not carry enough supplies to support a family for the journey
or at the destination. These pioneers brought tools with them
to be self-sufficient [Williams 2016]. Our restructuring relies
on self-replication, but more important, on the production of
tools to make the restructuring effort self-sufficient and sufficiently productive.
You can't just say making a "truss building unit" would take a spider 10550 hours of labor, or that spiders operating TBUs will produce truss 416 times faster - show me the design for this spring powered truss builder that takes precisely 3 hours and 3 minutes to produce a unit of truss; show me the spider that builds it, show me the steps along the process of the spider assembling the unit, show me where these numbers are coming from. It would be one thing if these were rough order of magnitude estimations, but the level of precision implies that not to be the case.
I disagree with your assessment of importance. A work like this strikes me as being primarily important as a "concept of operation", with sketched numbers from (presumably) sub-studies. In other words, my primary concern would be Does the whole thing hang together? Hard to say for such a vast project, so you need to start somewhere with many estimates and placeholders. I wouldn't shoot down the value based on missing references for sub-projects. Especially for what appears to be a solo-author work.
I do agree those would be helpful references. I've contacted the author asking for collateral materials and will post back if there's anything to share.
The validity of the concept of operation depends entirely on the validity of the assumptions that went into it. If, for example, the truss building unit can't actually make truss much faster than an appropriately designed spider working directly, or if the project would be done before the TBU could be completed, then suddenly the argument for spending a lot of resources building the TBU disappears. Yeah, you gotta start somewhere, but this isn't the place to start.
What's more fun is imagine pointing this thing at earth instead of an asteroid. Soon the whole crust of the earth is turned into paperclips, or trusses or whatever. oops.
This kind of concept is explored in the (science fiction) novel Seveneves [0] by Neal Stephenson, specifically the first half. Also explored: reshaping and vaporizing the ice on a comet with a nuclear reactor to create an impromptu propulsion device.
2312 by Kim Stanley Robinson features the idea heavily (the main character is a former "asteroid terrarium designer"). The Expanse series has quite a bit of it too.
Stephen Baxter's Phase Space. If those short stories, especially the ones with Reid Malenfant, grab you, then read the Manifold series. If you find yourself wishing it was more Alastair Reynolds, then the Xeelee Sequence.
Using automata is a bit silly and has problems. The biggest problems are vacuum welding and lack of lubrication. Vacuum welding can cause higher wear or even failure on metal components. Gears and cams absolutely need lubrication, which is liable to evaporation in a vacuum or requires exotic elements such as molybdenum.
Gears and cams also need to be machined, and machining processes do not work well in a vacuum for some of the same reasons mentioned above. That being said, one ton of microchips goes a long way. Why bother with automata, when simple vacuum tubes(in space, a vacuum tube is a couple funny shaped pieces of metal in close proximity) and magnetic amplifiers could be used to amplify signals from microcontrollers into something capable of doing useful work?
Just spitballing: in a use case like TFA, building pieces and structures in/around an asteroid, would they not be subject to fouling by dust and debris from the processes going on around them?
Can they build a demo on earth? Maybe reshape a small hill-side into something fun or pretty with "Autonomous Restructuring". That would be fun to watch.
> Can they build a demo on earth? Maybe reshape a small hill-side into something fun or pretty with "Autonomous Restructuring". That would be fun to watch.
If you are restructuring an asteroid, containment isn’t much of a problem; if you want to do it to an isolated site on earth, it is more of a concern.
It's not "exotic" in that it isn't used (it has major industrial Uses) but rather that it is one of the less common materials in the crust. It's less common than many of the "rare earth" elements and less common than uranium (Uranium is the 49th most common element while Molybdenum is the 58th depending on source https://en.wikipedia.org/wiki/Abundance_of_elements_in_Earth... ).
Because of the large scale and also because that would be an extra step. This methodology works best if the problem of construction in space has some solution.
The novel 2312, by Kim Stanley Robinson, explores this idea and has a fun chapter describing the process of hollowing out and terraforming an asteroid, using the ejecta to give it artificial gravity and a stable orbit in the solar system. Highly recommended.
This is a great example of the space fallacy, the idea that things that are impossible to do on Earth are somehow practical technologies for space. If it's all so easy to do, then go hollow out your local mountain; I promise it will get people's attention.
I'd settle for hollowing out an actual rubble pile as a demo. I'd be ecstatic, honestly, if you dumped any package of machines on a rubble pile and it autonomously, with only solar power, produced any machine more complicated than a bar of metal (and even that bar of metal would be no joke). Refining raw materials is hard. At least in space you'd be able to melt things in the open without them immediately oxidizing to hell.
Gravity is dubious as pro or con. It's honestly kind of handy sometimes, as a spatial organizing principle if nothing else.
I mean, FTL is probably a pipe dream. Nuclear fusion is an economic problem, AGI is on track to happen within our lifetime, biological immortality is in principle possible if you're not super strict about the definitions (i.e. can't beat the heat death of the universe with it), and self-replicating robots are quite literally what all of biology is.
They all represent the pinnacle of their respective branch of science.
By biological immortality I mean your body doesn't fail, or is fixable - you don't die of old age or random kidney failure.
You still die if someone cuts off your head.
> self-replicating robots are quite literally what all of biology is
By the same argument AGI is just same as human brain.
Many people think ChatGPT is so ompressive, that we are on the cust of AGI. I am convinced otherwise - chat gpt makes limitations of current AI approached very clear.
My best guess is that AGI will be one of the last items, maybe after FTL
> By biological immortality I mean your body doesn't fail, or is fixable - you don't die of old age or random kidney failure. You still die if someone cuts off your head.
That sounds entirely possible then. I'd wager the limits we'd hit would be memories - it's not known if our brains, in their current form, would be able to cope with hundreds of years of memories.
>> self-replicating robots are quite literally what all of biology is
> By the same argument AGI is just same as human brain.
Yes, but the main difference is that biology is merely too big for us to keep track of - it just has too many moving pieces that have been, from our perspective, heavily overfitted. Still, we've reached the level where we can reprogram some of the nanobots. In contrast, the topic of intelligence still has a lot of big mysteries.
Another way of looking at it is, we already have a well-mapped way of building nanotechnology: reprogramming the one that is around us, and is us. Plenty of little self-replicating programmable bots to pick from. But we're not at similar stage with poking in brains just yet.
> Many people think ChatGPT is so ompressive, that we are on the cust of AGI. I am convinced otherwise - chat gpt makes limitations of current AI approached very clear.
I agree. Though to me, it also revealed limitations of human cognition (or rather, it was already quite apparent at GPT-2 level; ChatGPT is only rubbing it in everyones' faces).
Have you ever felt that your own thinking, in many situations, is mostly cache lookups? That your speech and inner monologue (if you have one) both resemble a Markov chain, and your "self" mostly just observes and censors the output? I certainly did, quite a lot over the decades. I know others did, many obviously rejecting it as silly association. But what the recent LLMs show us is that maybe, just maybe, it's not silly at all. Which invites the question, if a lot of our cognition works this way, then just how much more complicated are the bits that don't?
(Another thing ChatGPT and the like are making apparent, is that AI risk isn't tied to reaching AGI. I think people kind of assumed it would take AGI taking off to end human civilization, but at this point I believe it's pretty clear that LLMs as a class of models are up to the task already - all they need is to have more memory, access to APIs that let them affect the real world and observe the results, and being run continuously; Reddit corpus already supplies the token association patterns that would let them end the world if given the means...)
> I think people kind of assumed it would take AGI taking off to end human civilization, but at this point I believe it's pretty clear that LLMs as a class of models are up to the task already
Absolutely, and in addition to that, people with zero grasp of the risks make decisions to trust AI with dangerous things -> money, vehicles, markets, justice, etc.
> Which invites the question, if a lot of our cognition works this way, then just how much more complicated are the bits that don't?
I this we could be mistaking quantity for the important.
Maybe other aspects of thinking, the ones we perceive as mundane, are actually impossible for a computer.
Consider mechanical tasks - a hundred years ago we had huge breakthroughs with engines - oh look, we can make crazy strong machines, we will soon automate all manual labour and blue collar jobs will be a thing of the past!
Well, turns out power was the easy part. A robot arm can lift a car, but getting precision took decades. So now we have power and precision, but try to get a robot arm (under computer control) to undo buttons on your T-shirt without ripping the T-shirt or your face. If a robot arm touches a piece of fabric, or you skin, it has no idea. They routinely kill people.
If you need non-repetitive dexterous movement with fine sensory input, that's basically beyond the cutting edge for today's robots. We can't automate sewing of clothes, cleaning of delicate items, etc.
I agree the stuff any five year old can do is well beyond current AI/robotic tech.
Let's focus on actual intelligence, though. What, exactly, would a 100-mind AGI be able to do that a 1-mind human can't? And what can a 1-mind human do that the 100-mind AGI can't?
My guess is that the 100-mind AGI is capable of building nearly perfectly accurate internal models of the 1-mind humans. So the AGI runs the model, and figures out the exact stimulations needed to get the 1-mind human to do what the AGI wants. You will not be able to resist.
The 1-mind human might be able to come up with physical theories that the AGI has no way to deduce. Maybe. (These physical theories may need experimental corroboration. If the AGI doesn't have access to the inner details of the latest ultra-LHC experiments, it may not be able to properly theorize.)
Cool but why send people to space? Keeping humans alive in space makes the problem so much more difficult. Just move all our industry into space and have the humans stay home.
Asteroids are indeed much more dangerous because no atmosphere exists to protect machines and human from space scraps and lights.
That said I think human need to be multi-planet to be able to survive big disasters. And asteroids as a small scale space station has its own benefits compared to planets or space crafts.
Because our robotics technology isn't anywhere near sufficient to operate autonomously without needing repairs, or to make decisions on the spot without human intervention.
>: Our asteroid restructuring approach can be compared to the effort of early pioneers. A wagon heading west in America in the 1800s could not carry enough supplies to support a family for the journey or at the destination. These pioneers brought tools with them to be self-sufficient [Williams 2016]. Our restructuring relies on self-replication, but more important, on the production of tools to make the restructuring effort self-sufficient and sufficiently productive.