You aren't applying that method correctly. The method is just making the assumption that you are observing a entity at a point in its lifetime that is randomly

You aren't applying that method correctly. The method is just making the assumption that you are observing a entity at a point in its lifetime that is randomly distributed with a uniform distribution. So there is a 75% chance that you are in the middle 75% of its lifetime (i.e. somewhere from 12.5% to 87.5% through its lifetime) and there is a 95% chance that you are in the middle 95% (somewhere from 2.5% to 97.5% through the lifetime). The US is currently 242 years old, so the 75% confidence interval for its total lifespan is 276-1,936 years for an expected end date of 2052-3712. The 95% confidence interval end dates are 2024-11,456. Applying it in reverse, the chance of it collapsing within my expected lifetime is 18%. That's not exactly comforting.