3 ms·

Extreme longtermism is hard to discern from believing in heaven and hell. E.g. the future is so important that almost anything can be justified in the near ter

Extreme longtermism is hard to discern from believing in heaven and hell. E.g. the future is so important that almost anything can be justified in the near term because of the far away ends, which often we won't even life to experience. As well often the longtermism speculation about possibly "living forever" and where "anything is possible" where "all needs are met" and we will be "traveling among the stars". That sounds very Biblically inspired to me. The fact unfortunately is the future is really hard to predict with any certainty. So justifying any significant suffering in the near term in service of something that is far in the future is often a fools errand (or worse.) Some of the black-and-white scenarios posited also are unrealistic. It isn't likely either AGI or humanity that survives -- this is just a sci-fi plot. Usually it isn't an us or them situation in the world, less you want to motivate a populace to war. What usually happens is that all possible futures are achieved in some fashion and they coexist and there is long-term complex competition and cooperation and evolution between them. For example, there are probably human colonies and probably AGI colonies and hybrid colonies and they trade with each other and do their own things. Some will expand faster than others, etc. Contrary to most popular sci fi literature that is human centric (StarTrek, Star Wars, The Expanse, Foundation, etc), if you think about it, AGI / robots are infinitely better suited to exploring and exploiting the galaxy compared to us - infinite lifespans, transmissible consciousness, no need for biological life support, repairability, more varied physicality, ability to survive high G/radiation, etc.

thfuran 3y ago
>What usually happens is that all possible futures are achieved in some fashion and they coexist and there is long-term complex competition and cooperation and evolution between them. For example, there are probably human colonies and probably AGI colonies and hybrid colonies and they trade with each other and do their own things. Some will expand faster than others, etc. What is your reference for what is usual in this context?
- bhouston 3y ago
  > What is your reference for what is usual in this context? Probably a wishy washy statement, I apologize. I was sort of partly making an analogy to evolution where everything is generally tried out in a random fashion and then what works tends to stick around. It is really hard to predict what will work ahead of time because context matters and it is always changing. What may work for a while may not work in the future because of changing context. I also do not mean all possible futures that one can ever think. I mostly mean speculative futures that are often talked about in sci fi books. My experience in my life is that technological inventions talked about regularly in sci fi tend to be achieved over the long run, although it is hard to predict exactly when.
  - thfuran 3y ago
    I think that's a particularly poor example, given that an estimated 99 percent of all species are already extinct.
    - bhouston 3y ago
      > I think that's a particularly poor example, given that an estimated 99 percent of all species are already extinct. Huh? Anyhow, it is really hard to predict evolution over the long-term because of the ever changing context (environment, other species, etc.). Generally things happen and are tried out, experimented with and what works sticks around. Those claiming to be able to predict the future concretely from many potential futures are probably wrong (and are also likely fans of Azimov's Foundations series.)
      - etrautmann 3y ago
        Exactly. This is what's so confusing to me. How can any intelligent person think that they can predict what will happen even 10 years out, let alone 100 or 1000. The chaotic dynamics of nonlinear systems renders the Foundation series laughably ungrounded, and seems to be overlooked by many today.
      - thfuran 3y ago
        What almost always happens in the context of evolution is that things are tried out and what works sticks around until something changes and the species goes extinct. Nearly every species that ever existed is already extinct. It's hard to predict exactly what will happen in the long term with evolution, but that doesn't mean "the species that are now populous will always remain so" is a good bet.

thfuran 3y ago

>What usually happens is that all possible futures are achieved in some fashion and they coexist and there is long-term complex competition and cooperation and evolution between them. For example, there are probably human colonies and probably AGI colonies and hybrid colonies and they trade with each other and do their own things. Some will expand faster than others, etc. What is your reference for what is usual in this context?

bhouston 3y ago
> What is your reference for what is usual in this context? Probably a wishy washy statement, I apologize. I was sort of partly making an analogy to evolution where everything is generally tried out in a random fashion and then what works tends to stick around. It is really hard to predict what will work ahead of time because context matters and it is always changing. What may work for a while may not work in the future because of changing context. I also do not mean all possible futures that one can ever think. I mostly mean speculative futures that are often talked about in sci fi books. My experience in my life is that technological inventions talked about regularly in sci fi tend to be achieved over the long run, although it is hard to predict exactly when.
- thfuran 3y ago
  I think that's a particularly poor example, given that an estimated 99 percent of all species are already extinct.
  - bhouston 3y ago
    > I think that's a particularly poor example, given that an estimated 99 percent of all species are already extinct. Huh? Anyhow, it is really hard to predict evolution over the long-term because of the ever changing context (environment, other species, etc.). Generally things happen and are tried out, experimented with and what works sticks around. Those claiming to be able to predict the future concretely from many potential futures are probably wrong (and are also likely fans of Azimov's Foundations series.)
    - etrautmann 3y ago
      Exactly. This is what's so confusing to me. How can any intelligent person think that they can predict what will happen even 10 years out, let alone 100 or 1000. The chaotic dynamics of nonlinear systems renders the Foundation series laughably ungrounded, and seems to be overlooked by many today.
    - thfuran 3y ago
      What almost always happens in the context of evolution is that things are tried out and what works sticks around until something changes and the species goes extinct. Nearly every species that ever existed is already extinct. It's hard to predict exactly what will happen in the long term with evolution, but that doesn't mean "the species that are now populous will always remain so" is a good bet.