Systemic/procedural meta-issues are the most important political isssues. The payoff for having a better decison-making system is larger than the payoff for making the right decision on almost any individual issue.

Another reason: in the past, revolutions have often succeeded only when lower-ranking officers in the military rebeled and joined the revolution, against the direction of the top brass. Soon warfare will be carried out by robots, and so military rebellion will not occur as easily. This will make revolutions much more difficult. At that point, the constitutional structures then in place will be crystalized, and difficult to change in the event of corruption or authoritarian takeover. We need to find ways to refine constitutional protections against corruption and authoritarianism now (finding ways to strengthen them while retaining enough flexiblity that the protections aren't dropped later due to security concerns), so that they will be in place by that time.