For two square matrices $A$ and $B$ of the same size, we have the identity $$ \det\left[ \begin{matrix} A & B \\ B & A \end{matrix} \right] \ = \ \det\left( A + B \right) \det( A - B ) \ . $$ This is supposed to hold even when the matrices are non-commuting ie. $A B \neq BA$.
In the case that $AB=BA$, I can see how the above follows from the identity $$ \det\left[ \begin{matrix} A & B \\ C & D \end{matrix} \right] \ = \ \det(D) \det\left( A - B D^{-1} C \right) $$ by taking $(C,D) \to (B,A)$ and mutiplying the argument of the determinants together, leads to $$ \det\left[ \begin{matrix} A & B \\ C & D \end{matrix} \right] \ \to \ \det(A) \det\left( A - B A^{-1} B \right) = \det\left( A^2 - B A^{-1} B A \right) = \det\left( A^2 - B^2 \right) = \det\left( A - B \right)\det\left( A + B \right) $$ but the above proof relies heavily on $AB=BA$.
How do you see this identity is true in the case that $AB\neq BA$?