I can't tell whether you are denying assumption 1 or 2.
Thanks for this.
Even if this argument is successful, there are debates over decision theory (evidential, causal, functional). Does an ideally rational agent intervene at the level of states, actions, or decision procedures?
If it's decision procedures, or something similar, functional decision theory can you get views that look quite close to Kantianism.
I'm y'all are thinking about this.