Sometimes, someone's sole reason for not implementing a plan or solution to a problem is that it has flaws -- or that it will yield some negative results. The problem with this approach is that it's all-or-nothing, so incremental improvement is essentially disallowed by it. It's important to remember that solutions and potential courses of action should be compared not absolutely but relatively to their alternatives, including the alternative that's already in place. If a particular course of action allows for the possibility of something going wrong, and the current plan also does, then your concern shouldn't be whether your new plan will solve all of your problems; it should be whether the new plan will be better than the current one.
1. If you're in a warzone and realize that it's almost certain that many people around you are about to die, this should not prevent you from saving the ones whom you know you can save. Saving three may not be as great as saving fifty, but it's still better than saving zero.
2. If you're standing on train tracks and a train is heading straight for you, but there are murderous thugs on both sides of the tracks, it would be foolish to continue to stand on the tracks simply because the thugs pose a risk to your safety.