We don't necessarily need to build a Dyson sphere to reach other planets and stars.
And if the Dyson sphere gets built by an aggressive singularity to the tune of Stross' Accelerando, then pretty much every human in that singularity's light cone is at serious risk. If that happens, we may have to broadcast human mind vectors out into the galaxy and hope a less awful singularity catches us and gives us a simulated home with reasonable Mind Rights.
Humans have a survival impetus to settle multiple planets and star systems. But a fast burn self improving Dyson swarm might easily create more risk than it alleviates. We might therefore choose to stick to more conservative technologies to hit the stars, even with all your assumptions in place.
Slow growth swarms might be safe if there's physical reasons that a fast burn Dyson sphere is impossible. But if we uncover the potential for fast burn, we could enter a "swarms race", warring to achieve and control the most effective and rapidly accelerating swarm. In this setup, each major political power tries to start a fast burn, because otherwise an enemy will achieve it first and wipe out your swarm. The result could be, for example, multiple runaway Dyson swarms, fighting each other without our control.
Now, saying all this, I still think it would be fun to build a self improving and self replicating swarm. I just don't know if it would be very safe, politically in the short term, and for the species in the long term.
And if the Dyson sphere gets built by an aggressive singularity to the tune of Stross' Accelerando, then pretty much every human in that singularity's light cone is at serious risk. If that happens, we may have to broadcast human mind vectors out into the galaxy and hope a less awful singularity catches us and gives us a simulated home with reasonable Mind Rights.
Humans have a survival impetus to settle multiple planets and star systems. But a fast burn self improving Dyson swarm might easily create more risk than it alleviates. We might therefore choose to stick to more conservative technologies to hit the stars, even with all your assumptions in place.
Slow growth swarms might be safe if there's physical reasons that a fast burn Dyson sphere is impossible. But if we uncover the potential for fast burn, we could enter a "swarms race", warring to achieve and control the most effective and rapidly accelerating swarm. In this setup, each major political power tries to start a fast burn, because otherwise an enemy will achieve it first and wipe out your swarm. The result could be, for example, multiple runaway Dyson swarms, fighting each other without our control.
Now, saying all this, I still think it would be fun to build a self improving and self replicating swarm. I just don't know if it would be very safe, politically in the short term, and for the species in the long term.