I've tried to solve this exercise based on a similar question that was asked some years ago, but I'm stuck in step 5. Any help? Thanks in advance. By the way I'm using Stanford's system.
1. AX:p(X) Premise
2. ~p(X) Assumption
3. EX:p(X) Assumption
4. ~p(X) Reiteration: 2
5. EX:p(X) => ~p(X) Implication Introduction: 3, 4
The idea is that you assume $\exists x \,\lnot P(x)$ and show that it leads to a contradiction with the premise $\forall x \, P(x)$, by means of the rules for the elimination of existential and universal quantifiers. This means that the negation of the assumption $\exists x \,\lnot P(x)$ holds.
This intuition is a little watered-down in Stanford's platform because its rules for negation introduction and existential elimination are slightly clumsy and unusual for natural deduction.
Anyway, the correct formalization of a proof of $\lnot \exists x \, \lnot P(x)$ from $\forall x \, P(x)$ in the Fitch-style natural deduction system implemented by Stanford's platform is the following: