Oh here we go -- this is a vast subject (and I understand some of your questioning), I wish I had time for a more thorough analysis, but here are some key point

Oh here we go -- this is a vast subject (and I understand some of your questioning), I wish I had time for a more thorough analysis, but here are some key points. - Humans are actually fantastic space exploration hardware. We already have robotic vehicles exploring other planets. They are not autonomous, and have glorified Arduinos in them. Not for lack of money or development, but because it is extremely difficult to do autonomous exploration, and the hardware is essentially currently impossible because there's nothing good enough, and not even remotely close to good enough that can survive radiation and space environment. Exploration currently is extremely slow, done through remote control, with up to 24 minute delays (Mars). - Humans can self-reproduce with low-tech inputs We are perhaps from >50 to >>100s years away from machine compact self-reproduction, then imagine self-reproduction with human-like intelligence. Only humans can make colonies for the near future. - AGI (or just good enough scientist AI, explorer AI, etc) is uncertainly distant It could be 20, or it could be 100 years or more away (if our civilization even survives that long before wars/climate change/etc), economically viable AGI is uncertain, economically viable compact radiation-resistant AGI is vastly uncertain. - It is worthwhile to bet on long term space exploration and habitation This is the most difficult case. I think it's worthwhile to devote resources on long term space exploration and habitation. In particular research since it is so expensive right now and we still experiencing rapid technological change. We want to eventually live on other planets (and even other star systems). The other aspects are complementary, like inspiring people to work in engineering and science; the main thing is giving hope for the future, even if it's a future you won't live. Most of the research anyways finds applications in terrestrial systems (remember that we still do pure Mathematics, whose research value tends to crop decades later or more). --- It is expensive, yes (ISS costs about $3 billion per year for NASA apparently), but overall I think it's a tiny investment for great returns (at varying levels of risk). --- Finally, to expand on hope: hope is great things many things, but one of them is simply the way people act on expectations of the future. If they don't expect to live long, or a long term future worth living for other generations, you expect we will act more selfishly and greedily, a self-fulfilling prophetical aspect. Thus it's extremely important not to neglect sustainability and somewhat distant dreams.