Regarding @SaalHardali's question, and @DenisNardin's answer, I think this can be generalized and maybe simplified.
First of all, we can generalize from Cat to any SM category C, and ask if an invertible morphism in CAlg^lax(C) is in fact in CAlg(C).
We can use the model Fun^x(Span(Fin), C) to model CAlg(C) and CAlg^lax(C), where the morphisms are natural transformations and lax natural transformations respectively.
So I think that then the question becomes: is a lax natural transformation which is invertible a (strict) natural transformation.