Closedness of the derived-wild locus in algebraic families

Let $\mathcal A$ be a family of finite-dimensional algebras based on an algebraic variety $X$, and let

$$W=\{x\in X\mid \mathcal A(x)\text{ is derived wild}\}.$$

Closedness conjecture. In the situation of Corollary 44, the set $W$ is always closed in $X$; equivalently, the set

$$\{x\in X\mid \mathcal A(x)\text{ is tame}\}$$

is open. This would strengthen the preceding result that the derived-wild locus is a countable union of closed subsets, giving a genuine open–closed dichotomy for tame and wild fibres in algebraic families. The source does not provide a resolution of this conjecture.