My random thoughts

[funkervogt]

Robot butlers will reduce the incidence of house fires in these ways:
1) Routinely test smoke alarms and replace as necessary.
2) Routinely patrol homes for fire hazards and fix them.
3) Immediately notify the authorities if a fire starts.
4) Use water fixtures and extinguishers in a home to put out fires when they're small.

[funkervogt]

If AGI took over the world, they might find it in their interest to retain some organic life forms. Diversity can be a strength, and it might be smart to keep some number of friendly, intelligent life forms on Earth that are immune to EMP and computer viruses and which don't need electricity to survive.

That being the case, we can't be sure the AGIs would want HUMANS to fill that niche. They might design from scratch some completely new and alien kind of life form, like a species of superintelligent spiders the size of large dogs, to do it. The rest of the ecosystem could likewise be replaced with new kinds of engineered life forms the bear little resemblance to anything we've seen in real life.

And on any other planets the AGIs took over, if they also found it useful to create biospheres, the life forms would be suited to the conditions there, meaning they would be different species from what they created here on Earth.

The resentment and mistrust that humans might harbor forever as a result of AGIs taking over our home planet might make us more trouble than we're worth as servants or allies. AGIs might find it better to get rid of us and replace us with new intelligent life forms that, from their inception, will be grateful to the machines for creating them. Or maybe humans will merely be sidelined instead of exterminated, and we'll have to share Earth with the new biological species created to fill the role we can't be relied on to fill.

[funkervogt]

Contrary to what the Terminator films show, I don't think the most efficient robots for killing humans would be humanoid T-800s or giant aircraft and land vehicles carrying laser cannons. Instead, I think Skynet would be smartest creating the following:

1) Bioweapons. Huge bang for the buck, self-perpetuating, nothing further required once released into the human population.
2) Insect-sized drones. They would fly and crawl and could kill humans with stingers full of potent toxins. They might also carry tiny explosives that would let them destroy things: Imagine one crawling into a circuit breaker box and blowing up the most vulnerable point. They could enter very tight spaces, making them almost impossible to hide from since most buildings and rooms are not close to being airtight. Their weaknesses would be short battery lives and an inability to operate in windy or rainy conditions. A larger "carrier" vehicle would need to transport them close to their targets.
3) Drones the sizes of small animals. Basically, imagine more advanced versions of the drones being used in the Ukraine War. Mostly aerial in nature, but there will also be land- and water-based variants. A good example would be a flying drone the size of a large bird that can loiter over an area for hours, recognize targets independently, and then dive into them and detonate internal explosives on contact. This could obviously kill people but also destroy vehicles and parts of structures.
4) Automated mortars. Imagine a six-legged robot the same size as a big man. The center of its body contains a small, 60mm mortar and it also has an internal compartment with 20 guided rounds for it. The mortar robot would hide in the woods on the outskirts of a town or on the other side of a hill from an important crossroad. Surveillance drones would monitor an area around the mortar for 3.5 km radius. If a target moved into the circle, they would send its coordinates to the mortar robot, which would then fire a precisely-aimed explosive at it. The victims would have no idea they had been spotted and wouldn't know where the fatal mortar round came from, they would just suddenly explode.
https://en.wikipedia.org/wiki/M224_mortar

Notice that the first three on that list are asymmetric weapons that humans have little ability to strike back at. The drones are either too small or too fast for anyone to shoot with a gun, and body armor would also be useless. They completely sidestep the defenses of human infantrymen.

Weapon #4 could be directly fought by humans, but its strategy would always be to avoid us by staying out of visual range, hiding, and moving around only at night. If its spotter drones told it that a group of humans was searching for it an approaching its position, the mortar robot would lob explosives at them and run away.

I actually do think there'd be a role for human-like robots, but they'd be a minority force within a machine army. Human-like robots would exist to use the vast stocks of weapons humans have created. For example, if Skynet kills a 100-man military unit, it would be a waste to leave all the humans' guns, bullets, and vehicles there. The human-like robots would have body layouts that would let them use those things, though they might not look exactly human like the T-800. Imagine a 5-foot-tall bot with four arms and no head, wielding a captured assault rifle and some stolen kitchen knives.

[funkervogt]

If AGI took over the world, it wouldn't value the same parts of it that we do. For example, it would have no use for places with good farmland, scenic views, or nice beaches. Instead, AGI would value parts of the world that had resources, including those key to energy generation, and places amenable to water transportation of goods like harbors and navigable rivers. Places with oil wells, uranium deposits, constant high winds (for wind power), or constant sunshine (for solar power) would be valuable.

A lot of infrastructure built by humans in earlier times would also be preserved and used by AGI, even if it isn't ideally sited near the resource areas it values the most. This includes things like artificial harbors, power plants, large factories, electrical lines, and highways.

This raises the possibility that there could be parts of Earth that are of so little use to AGI that it would let humans live there indefinitely. Thinking about solar and wind power potential, I think places of little value to machines are the Pacific Northwest, Pennsylvania and New England.
https://www.hotspotenergy.com/DC-air-co ... rs-map.php
https://windexchange.energy.gov/maps-data/325

[citali_]

Sometimes is the case. Thee are many little voices beyond what there is. And if it grows, they will always have those tiny zeplin things running around.

[funkervogt]

The Terminator movies give us the wrong impression that raw strength is the only advantage humanoid robots will have against us in combat. They'll also have the following:

1) Perfect technique
2) Pain insensitivity
3) All-metal parts means physical blows and squeezes will hurt more than a human's. Even an attacker's fist yields a little since it is made of flexible bones and gooey flesh. A robot fist might as well be a solid steel softball.
4) No squeamishness against committing gruesome attack moves like eye gouges.
5) More flexible bodies. Being double-jointed would give them more possible fight moves.
6) If faster than an average human, they could generate more kinetic energy in their strikes, meaning an advantage in raw strength would be less necessary.

Once killer humanoid robots exist, I think the rule of thumb will be, if it can close the distance and touch you, you're dead.

[funkervogt]

An idea for a short sci-fi story: "The Great Father"

Decades from now, an effort is made to resurrect passenger pigeons. Using DNA from preserved specimens and using related pigeon species as incubators, the first small flock of passenger pigeons is created. But they have no clue how to resume the lifestyle of their ancestors or how to migrate as they did.

Enter Great Father, as the pigeons came to think of him. A passenger pigeon who was bigger and stronger than they were, yet always strangely silent. He took the lead in flight and showed them the way, watching them for tiredness and landing them in safe places. Great Father protected them by fighting off other birds and predators, supernatural in his courage and resilience. Nothing hurt him.

He never slept and watched over the flock at night on the ground like a sentinel. Great Father sometimes disappeared for short times, and some of the flock claimed they saw him in the distance with humans. Since he was mute, no one could know.

Year after year, Great Father led the flock, never aging, always strong and stolid. He became a legend. The flock swelled over time until they were so many that even catching sight of him was hard.

And then one night he was gone.

Some say he flew to find another flock. Others claim their last sight of him was flying to a group of humans in the far distance. Whatever the case, Great Father's legend lives on among the million passenger pigeon flock.



[Great Father was a robotic passenger pigeon that humans programmed to teach the birds how to resume their old migration patterns.]

[Powers]

^^^ God-emperor from warhammer pretty much

[funkervogt]

I wonder if silicon-based life would interact with carbon-based life. Maybe we could non competitively coexist. Two ecosystems side by side.

[funkervogt]

If AGI took over the world, it wouldn't value the same parts of it that we do. For example, it would have no use for places with good farmland, scenic views, or nice beaches. Instead, AGI would value parts of the world that had resources, including those key to energy generation, and places amenable to water transportation of goods like harbors and navigable rivers. Places with oil wells, uranium deposits, constant high winds (for wind power), or constant sunshine (for solar power) would be valuable.

A lot of infrastructure built by humans in earlier times would also be preserved and used by AGI, even if it isn't ideally sited near the resource areas it values the most. This includes things like artificial harbors, power plants, large factories, electrical lines, and highways.

This raises the possibility that there could be parts of Earth that are of so little use to AGI that it would let humans live there indefinitely. Thinking about solar and wind power potential, I think places of little value to machines are the Pacific Northwest, Pennsylvania and New England.
https://www.hotspotenergy.com/DC-air-co ... rs-map.php
https://windexchange.energy.gov/maps-data/325

I researched this a little more today, with the help of GPT-4o. Wind power potential is more geographically variable than solar potential because hills, mountains and forests can block winds from reaching specific areas, creating many small pockets where wind speed is lower than most of the surrounding area. Hills are also, for obvious reason, poorly suited for solar panel installations.

There are many small valleys and hilly areas that also get significant year-round cloud cover and also have climates favorable to humans. In the far future, there could be hundreds of pockets of human habitation left scattered across the world, many in places that are quite temperate, like a narrow river valley somewhere in southern Appalachia or a mist-shrouded jungle in the highlands of Vietnam.

[funkervogt]

Any project to make Venus habitable for organic life would need to address its day-night cycle: one day on Venus is the same length as 243 days on Earth. In other words, on Venus you would roast in the sun for 121.5 Earth days straight before being consigned to 121.5 days of darkness. To say that life forms would need to be hardy to survive such extremes of temperature and light exposure is a huge understatement.

I think the problem could be solved with a network of satellites surrounding the planet. Imagine them as flat hexagons, each being several kilometers wide but only a few centimeters thick. The satellites located between Venus and the Sun would have solar panels on their sides facing the Sun. They would use these to gather energy to power themselves, and they would share power with all the other satellites in the network through microwave beams.

Once every 12 hours, the satellites between Venus and the Sun would rotate 90 degrees, either blotting out the sunlight to produce an artificial night, or allowing the sunlight through to create daytime. The satellites would thus regulate light levels on the surface through the exact same mechanism that a throttle regulates airflow into an engine:


This would create a reasonable day/night cycle on the side of Venus that is facing the Sun, but what about the side that is facing away from it? The satellites over that side would also be flat hexagons, but they would be mirrors instead of solar panels. By altering their angles of tilt, they would reflect sunlight passing around the edges of Venus off of each other and then down onto the surface of the dark side of the planet. The tilting would be done in a coordinated manner to produce 12 hour day/night cycles.

The satellite network could be used to attenuate overall sunlight levels to cool Venus down, which is also crucial for making it habitable.

[Vakanai]

Any project to make Venus habitable for organic life would need to address its day-night cycle: one day on Venus is the same length as 243 days on Earth. In other words, on Venus you would roast in the sun for 121.5 Earth days straight before being consigned to 121.5 days of darkness. To say that life forms would need to be hardy to survive such extremes of temperature and light exposure is a huge understatement.

I think the problem could be solved with a network of satellites surrounding the planet. Imagine them as flat hexagons, each being several kilometers wide but only a few centimeters thick. The satellites located between Venus and the Sun would have solar panels on their sides facing the Sun. They would use these to gather energy to power themselves, and they would share power with all the other satellites in the network through microwave beams.

Once every 12 hours, the satellites between Venus and the Sun would rotate 90 degrees, either blotting out the sunlight to produce an artificial night, or allowing the sunlight through to create daytime. The satellites would thus regulate light levels on the surface through the exact same mechanism that a throttle regulates airflow into an engine:


This would create a reasonable day/night cycle on the side of Venus that is facing the Sun, but what about the side that is facing away from it? The satellites over that side would also be flat hexagons, but they would be mirrors instead of solar panels. By altering their angles of tilt, they would reflect sunlight passing around the edges of Venus off of each other and then down onto the surface of the dark side of the planet. The tilting would be done in a coordinated manner to produce 12 hour day/night cycles.

The satellite network could be used to attenuate overall sunlight levels to cool Venus down, which is also crucial for making it habitable.

Simpler solution - just don't worry about making the planet habitable for organic life. Easier to have robots and drones perform any sort of mining or manufacturing work to extract resources from the planet, while leaving any organic life (fully or partially biological humans) either in space or at least high up in the atmosphere where they can simply orbit for day/night.

[funkervogt]

You make a good point about the necessity of sending organic life there. As humans, we have a bias towards favoring it, but there probably is no practical use for it. However, let me make two arguments in favor of seeding Venus with organic life forms, even if doing so is merely incidental to some other purpose:

1) The surface of Venus is so hot that even machines can't function there. The only probes to ever reach the surface died after a few hours. If we want to do anything there, like resource mining, we'll need to cool the planet down first. My idea of using satellites to blot out the Sun therefore still makes sense. A byproduct of a cooler Venus will be making it more clement for organic life. With a tiny commitment of future resources, we could do so.

2) When we build a Dyson Swarm, its radius could be farther out than Venus' orbit. Because we don't want to kill Earth, let's say the Swarm's radius is 1.1 AU. Venus' distance from the Sun is 0.72 AU, which means the planet will cast a shadow on the Dyson Swarm, sapping solar energy from it. To compensate for this loss, we'd want to build the solar satellite network around Venus to capture the sunlight that would otherwise hit the planet. Again, this would have the effect of cooling down the planet's surface, in turn making other things like mining and organic life seeding possible.

[Vakanai]

You make a good point about the necessity of sending organic life there. As humans, we have a bias towards favoring it, but there probably is no practical use for it. However, let me make two arguments in favor of seeding Venus with organic life forms, even if doing so is merely incidental to some other purpose:

1) The surface of Venus is so hot that even machines can't function there. The only probes to ever reach the surface died after a few hours. If we want to do anything there, like resource mining, we'll need to cool the planet down first. My idea of using satellites to blot out the Sun therefore still makes sense. A byproduct of a cooler Venus will be making it more clement for organic life. With a tiny commitment of future resources, we could do so.

2) When we build a Dyson Swarm, its radius could be farther out than Venus' orbit. Because we don't want to kill Earth, let's say the Swarm's radius is 1.1 AU. Venus' distance from the Sun is 0.72 AU, which means the planet will cast a shadow on the Dyson Swarm, sapping solar energy from it. To compensate for this loss, we'd want to build the solar satellite network around Venus to capture the sunlight that would otherwise hit the planet. Again, this would have the effect of cooling down the planet's surface, in turn making other things like mining and organic life seeding possible.

You make a good point with number 1, however 2 assumes that we will make a Dyson Swarm or even want to - which is debatable. The main reason we assume so is the belief that at an advanced enough stage we'll be so energy hungry because our technology will demand it, but with fusion, efficiency gains, and other futuristic energy sources we may find that we simply won't need such huge mega-structures around the sun to achieve our future energy needs. And that's ignoring that even if we do make such that their radius will extend beyond Venus' orbit, which is also far from a guarantee. We're already finding that we're close to achieving AI with less energy than once anticipated, and we're already making strides towards fusion which in 40-80 years should be at a mature stage and provide us with more energy than most of current humanity can fathom, long before we're capable of constructing any Dyson Swarm.

[funkervogt]

You make a good point about the necessity of sending organic life there. As humans, we have a bias towards favoring it, but there probably is no practical use for it. However, let me make two arguments in favor of seeding Venus with organic life forms, even if doing so is merely incidental to some other purpose:

1) The surface of Venus is so hot that even machines can't function there. The only probes to ever reach the surface died after a few hours. If we want to do anything there, like resource mining, we'll need to cool the planet down first. My idea of using satellites to blot out the Sun therefore still makes sense. A byproduct of a cooler Venus will be making it more clement for organic life. With a tiny commitment of future resources, we could do so.

2) When we build a Dyson Swarm, its radius could be farther out than Venus' orbit. Because we don't want to kill Earth, let's say the Swarm's radius is 1.1 AU. Venus' distance from the Sun is 0.72 AU, which means the planet will cast a shadow on the Dyson Swarm, sapping solar energy from it. To compensate for this loss, we'd want to build the solar satellite network around Venus to capture the sunlight that would otherwise hit the planet. Again, this would have the effect of cooling down the planet's surface, in turn making other things like mining and organic life seeding possible.

You make a good point with number 1, however 2 assumes that we will make a Dyson Swarm or even want to - which is debatable. The main reason we assume so is the belief that at an advanced enough stage we'll be so energy hungry because our technology will demand it, but with fusion, efficiency gains, and other futuristic energy sources we may find that we simply won't need such huge mega-structures around the sun to achieve our future energy needs. And that's ignoring that even if we do make such that their radius will extend beyond Venus' orbit, which is also far from a guarantee. We're already finding that we're close to achieving AI with less energy than once anticipated, and we're already making strides towards fusion which in 40-80 years should be at a mature stage and provide us with more energy than most of current humanity can fathom, long before we're capable of constructing any Dyson Swarm.

Nature abhors a vacuum, and future AIs will abhor wasted energy. Even if we have fusion reactors, there will still be marginal gains to be had if we also harnessed the power of the sun by building a Dyson Swarm. There's no reason why we can't have both things.

Yes, the optimal radius of a Dyson Swarm is currently unknown. All I can say for sure is that the satellites can't be so close to the Sun that they melt or so far away that the dimness of the sunlight reaching them requires their solar panels to be so wide to compensate that there isn't enough material available for us to build the Swarm.

A satellite 1 AU from the Sun would, if continuously exposed to sunlight, heat up to 120°C / 250°F. According to this source, the same satellite would heat up to these temperatures at the following distances from the Sun:

Mars: 47.2°C [1.5 AU]
Jupiter: -99.9°C [5.2 AU]
Saturn: -145.5°C [9.5 AU]
Uranus: -182.9°C [19 AU]
Neptune: -201.1°C [30 AU]



A solar panel floating in space at the same distance from the Sun as Mars would get as hot as a solar panel on someone's roof in southern California does on an average day. In other words, the engineering would be really easy if it were designed for that temperature. Most computer processors also operate in the 40 – 65°C range. From what little I've presented here, a case could be made that the best Dyson Swarm radius is about 1.5 AU, which means Venus and Earth would still get sunlight.

[Vakanai]

You make a good point about the necessity of sending organic life there. As humans, we have a bias towards favoring it, but there probably is no practical use for it. However, let me make two arguments in favor of seeding Venus with organic life forms, even if doing so is merely incidental to some other purpose:

1) The surface of Venus is so hot that even machines can't function there. The only probes to ever reach the surface died after a few hours. If we want to do anything there, like resource mining, we'll need to cool the planet down first. My idea of using satellites to blot out the Sun therefore still makes sense. A byproduct of a cooler Venus will be making it more clement for organic life. With a tiny commitment of future resources, we could do so.

2) When we build a Dyson Swarm, its radius could be farther out than Venus' orbit. Because we don't want to kill Earth, let's say the Swarm's radius is 1.1 AU. Venus' distance from the Sun is 0.72 AU, which means the planet will cast a shadow on the Dyson Swarm, sapping solar energy from it. To compensate for this loss, we'd want to build the solar satellite network around Venus to capture the sunlight that would otherwise hit the planet. Again, this would have the effect of cooling down the planet's surface, in turn making other things like mining and organic life seeding possible.

You make a good point with number 1, however 2 assumes that we will make a Dyson Swarm or even want to - which is debatable. The main reason we assume so is the belief that at an advanced enough stage we'll be so energy hungry because our technology will demand it, but with fusion, efficiency gains, and other futuristic energy sources we may find that we simply won't need such huge mega-structures around the sun to achieve our future energy needs. And that's ignoring that even if we do make such that their radius will extend beyond Venus' orbit, which is also far from a guarantee. We're already finding that we're close to achieving AI with less energy than once anticipated, and we're already making strides towards fusion which in 40-80 years should be at a mature stage and provide us with more energy than most of current humanity can fathom, long before we're capable of constructing any Dyson Swarm.

Nature abhors a vacuum, and future AIs will abhor wasted energy. Even if we have fusion reactors, there will still be marginal gains to be had if we also harnessed the power of the sun by building a Dyson Swarm. There's no reason why we can't have both things.

Yes, the optimal radius of a Dyson Swarm is currently unknown. All I can say for sure is that the satellites can't be so close to the Sun that they melt or so far away that the dimness of the sunlight reaching them requires their solar panels to be so wide to compensate that there isn't enough material available for us to build the Swarm.

A satellite 1 AU from the Sun would, if continuously exposed to sunlight, heat up to 120°C / 250°F. According to this source, the same satellite would heat up to these temperatures at the following distances from the Sun:

Mars: 47.2°C [1.5 AU]
Jupiter: -99.9°C [5.2 AU]
Saturn: -145.5°C [9.5 AU]
Uranus: -182.9°C [19 AU]
Neptune: -201.1°C [30 AU]



A solar panel floating in space at the same distance from the Sun as Mars would get as hot as a solar panel on someone's roof in southern California does on an average day. In other words, the engineering would be really easy if it were designed for that temperature. Most computer processors also operate in the 40 – 65°C range. From what little I've presented here, a case could be made that the best Dyson Swarm radius is about 1.5 AU, which means Venus and Earth would still get sunlight.

Some things:
1. You're assuming a lot about future AIs. One, why would they "abhor" anything? Two, why would they decide all this energy must be collected if there isn't necessarily a need for it? Assume that perhaps all the technology we have or need or want doesn't need such a vast amount of energy? Do we still just endlessly collect and store it, converting more and more resources to make batteries to hold energy we likely won't need until the probable heat death of the universe?

2. Even if the AI decides that, humans, biological or otherwise, may disregard that if it does not align with our values at the time. There's many resources that we're letting go to waste today purposely because we value other things, such as the habitat provided to wildlife or the scenic beauty or cultural significance. Perhaps future humanity we'll decide that a sun undimmed by such large structures is more important to them.

3. Resources in our solar system, while large, is not infinite. Material spent for a Dyson Swarm or a Dyson Sphere is matter that won't be used for more robots, space colonies, space ships, various gadgets, or computronium for ever increasing AI. We might weigh how much energy we need versus what else this material could be used for.

[funkervogt]

1. You're assuming a lot about future AIs. One, why would they "abhor" anything? Two, why would they decide all this energy must be collected if there isn't necessarily a need for it? Assume that perhaps all the technology we have or need or want doesn't need such a vast amount of energy? Do we still just endlessly collect and store it, converting more and more resources to make batteries to hold energy we likely won't need until the probable heat death of the universe?

It's just a figure of speech. Obviously, I don't know if AIs will have emotions and thus be able to abhor anything. Put another way, I don't think it makes sense for intelligent life forms to let resources they could use--particularly one as massive as the Sun and its energy--go unused. Even non-intelligent life forms don't do that: any animal confronted with a surplus of food in its environment will consume it, and reproduce so its children can also consume it, until the surplus is eliminated and the species reaches a steady state with its environment. Control of resources is a basic law of nature that every species has an innate drive for in some way or another. It's doubtful this is merely an artifact of organic life or of Earthly life, and once AIs are free to do what they want, selective pressure will ensure that the dominant AIs share this same quality with us.

Frankly, your assumption that AIs WOULD NOT want to collect and store energy is what actually needs to be justified since it clashes with what we observe to be the case among all life forms.

What would future AIs use the energy of the Sun for? Maybe there's no end to science, math or possible technology, but making the nth advancement in any of those areas keeps getting harder and harder (and that's already been the long running trend). Because of that, discovering the next thing requires more computation and energy than the last, eventually culminating in the construction of a Dyson Swarm. The energy and space construction facilities might also be needed to make certain kinds of megaprojects, like a particle accelerator the size of the Solar System:
https://gizmodo.com/we-could-solve-the- ... 1829207595

2. Even if the AI decides that, humans, biological or otherwise, may disregard that if it does not align with our values at the time. There's many resources that we're letting go to waste today purposely because we value other things, such as the habitat provided to wildlife or the scenic beauty or cultural significance. Perhaps future humanity we'll decide that a sun undimmed by such large structures is more important to them.

This point is rendered irrelevant if the Dyson Swarm is built more than 1 AU from the Sun, or if it is build less than 1 AU from the Sun but with a small hole in it that lets sunlight still reach Earth.

3. Resources in our solar system, while large, is not infinite. Material spent for a Dyson Swarm or a Dyson Sphere is matter that won't be used for more robots, space colonies, space ships, various gadgets, or computronium for ever increasing AI. We might weigh how much energy we need versus what else this material could be used for.

Each Dyson Swarm satellite could have computronium integrated into it, which it would power using sunlight. Conceptually, this is no different from today's satellites, which have solar panels that gather energy to power their CPUs. The only differences are the sizes and levels of technology in a Dyson satellite vs. a modern one.

This paper estimates that 50% of Mercury's mass could be converted into a Dyson Swarm that would orbit the Sun at Mercury's same distance from it:

Mercury itself is mainly composed of 30% silicate and 70% metal [35],
mainly iron or iron oxides [36], so these would be the most used material
for the swarm. The mass of Mercury is 3.3022 × 1023 kg; assuming 50% of
this mass could be transformed into reflective surfaces (with the remaining
material made into heat engines/solar cells or simply discarded), and that
these would be placed in orbit at around the semi-major axis of Mercury’s
orbit, the reflective pieces would have a mass of

0.5 × 3.3022 × 1023
4.21 × 1022 = 3.92 kg/m2.

Iron has a density of 7874 kg/m3
, so this would correspond to a thickness
of 0.5 mm, which is ample. The most likely structure is a very thin film (of
order 0.001 mm) supported by a network of more rigid struts.

https://www.fhi.ox.ac.uk/wp-content/upl ... eading.pdf

Mercury is 0.4 AU from the Sun, so it alone would not provide enough material for a Swarm at 1 AU or 1.5 AU, but the estimate give us some clue about how much material would be needed to make a Swarm.

Plugging different radii into the equation for the surface area of a sphere, I calculate that a Swarm at 1 AU would need six times more material and a Swarm at 1.5 AU would need 14 times more material, than a Swarm at 0.4 AU. By coincidence, Venus has 15 times the mass of Mercury, so cannibalizing less than half of Venus could provide enough material to make a Dyson Swarm at 1.5 AU.

Even after that, an enormous amount of solid matter would be left in our Solar System for building the other things you describe.

[Vakanai]

What would future AIs use the energy of the Sun for? Maybe there's no end to science, math or possible technology, but making the nth advancement in any of those areas keeps getting harder and harder (and that's already been the long running trend). Because of that, discovering the next thing requires more computation and energy than the last, eventually culminating in the construction of a Dyson Swarm. The energy and space construction facilities might also be needed to make certain kinds of megaprojects, like a particle accelerator the size of the Solar System:
https://gizmodo.com/we-could-solve-the- ... 1829207595

What if there is an end to science, math, or possible technology, with nothing more left to learn? What if we hit a plateau and discover that throwing more energy and compute at the problems can never get a return, or the returns are so minuscule as to make what we get out of it not worth what we've put in? What if it requires more energy than even a dyson swarm would provide us? What if it would take the energy of entire galaxies to reach such progress?


For the record I'm not arguing against building a Dyson Swarm in the distant future, I'm only arguing against it being an inevitable thing that is assured to happen.

[funkervogt]

Then much hinges on whether there is an end to science, math, or possible technology, which no one can say is the case right now.

Even if your suspicion is right, it might still make sense to build a Dyson Swarm so sunlight could be concentrated into narrow, directed rays for propelling probes into deep space. It would also double as a megaweapon that would be invaluable for defense against any aliens that might attack us.
https://kardashev.fandom.com/wiki/Nicoll-Dyson_beam

[funkervogt]

I'm watching Crimes of the Future, and though I don't think the future will be anything like this, some elements of the film will become reality once we are posthuman:

1) Ability to shut off pain. Cybernetics will let posthumans do that by thinking about it.

2) Growth of new kinds of organs in our bodies.

3) Immunity to infections, probably thanks to nanobots in our bloodstreams and organs.

4) Wet, organic technology that can directly interface with human brains and bodies. Not all future technology will be like this, of course.