We can actually make petty good estimates because of things like carbon layers in the ice. It's happened in the past, you're right, and usually it precedes large scale extinction events.
But when making actual predictions, the models need to make assumption about anthropogenic parameters like CO2 and methane emissions. That's the largest remaining uncertainty at this point. Given the same assumptions, climate models generally agree on the outcome.
Science is best when it’s purely that, I’ve seen plenty of living examples and read about past ones where science mixed with politics or overt profit motives don’t end well. Surely there must be examples where the contrary has been the case, but I am biased, and I would wager that it ended poorly more often than well.
I would much rather have politicians that heed scientific results than scientists springboarding into politics.