Future of Humanity Institute
1 HN recommender · recommended in 6 months, 2014–2026
48:07 · 46K views · Feb 1, 2012
Stuart Armstrong of the Future of Humanity Institute gives an Oxford physics talk on what exploratory engineering suggests about the Fermi paradox. He argues that von Neumann probes and Dyson spheres are plausibly buildable and asks what traces advanced civilizations would leave.
Cited as a source for the feasibility of colonizing the universe with self-replicating probes and for building a Dyson swarm from Mercury.
6 comments from Hacker News
If you found the above clip interesting and would like to learn more about the implications of such craft I highly recommend this video: this video video. The presentation covers the viability of constructing Von Neumann probes for colonizing the entire universe and illustrates the feasibility of it with a few different technological scenarios. All in all it seems feasible and the implications of that are enormous.
> I'm not sure which definition of 'growth' you're referring to here. One in which brain uploads are relevant, because you mentioned it. Literally any kind of efficiency has a maximum, an economy of pure minds was just the most extreme case which I took specifically to avoid any possible debate: there is (according to currently known laws of reality) definitely an absolute maximum in this universe, no matter what we do. Limit the population to current number or lower, and any other kind of efficiency also has a limit — lights can’t convert less than 0% into heat, resistance can’t be less than in a superconductor, reversible computers can’t beat the Landauer limit. You might fully automate manufacturing and give everyone alive today their own personal galaxy cluster [0] and it’s still finite. Big, still finite. > If it's GDP (which the article linked earlier did), well, in many developed countries, population has been decreasing despite increasing GDP. So this issue does not prevent growth. Indeed it is decreasing. Yet you may have to enforce that, because subgroups that grow faster rapidly dominate. This is unavoidable unless you take active measures to prevent it, because that’s what evolution does. (You can because evolution is dumb, but you have to). (I’d argue that in such a case, the license to reproduce would constitute a currency and in that currency humanity today world seem unimaginably wealthy; however this is not something I am prepared to make a keystone argument)
2100 is way too far out to predict anything . The cost of gene synthesis is roughly 20 years behind gene sequencing [0]. Extrapolated, by 2040, arbitrarily modifying your offspring will represent a trivial fraction (~0.5%) of the total cost of starting a family. Follow Moore’s law (considering lower costs rather than boosted performance) and assuming the Kurzweil estimate for full brain emulation, then by 2047, uploaded minds would be cheaper than the food eaten by subsistence farmers. [1] The first thing most people suggest when they hear about 3D printers is printing another 3D printer. It is definitely possible because that’s what every living cell does; while nobody knows how to do this now , if we can make it happen with the rocks of Mercury, it only takes four decades to build a Dyson swarm. [2] The questions are, what happens first, and how does humanity collectively react to that? [0] [LINK] [1] [LINK] [2] this video
Space is big, but Helios isn’t us trying to go as fast as possible, it’s going the speed it is because that’s a requirement of orbital dynamics. If we really really wanted, nuclear options could get us to a few percent of c with current tech and no breakthroughs. On the other hand, if someone builds a von Neumann probe that eats sunlight and Mercury (the planet), it could make a Dyson swarm in a human lifetime, and that would enable us to colonise most of our causal future light cone directly from Sol: this video
Kurzgesagt didn't invent the concept of disassembling Mercury to build a Dyson swarm. Stuart Armstrong proposed it in a lecture in 2012[0]. [0]: this video