Increasing current interest in sponge fossils includes numerous reports of diverse vermicular and peloidal structures interpreted as keratose sponges in Neoproterozoic to Mesozoic carbonates and in various open marine to peritidal and restricted settings. Reports of their occurrence are fundamental and far-reaching for understanding microfacies and diagenesis where they occur; and fossil biotic assemblages, as well as wider aspects of origins of animals, sponge evolution/ecology and the systemic recovery from mass extinctions. Keratose sponges: 1) have elaborate spongin skeletons but no spicules, thus lack mineral parts and therefore have poor preservation potential so that determining their presence in rocks requires interpretation; and 2) are presented in publications as interpreted fossil structures almost entirely in two-dimensional (thin section) studies, where structures claimed as sponges comprise diverse layered, network, particulate and amalgamated fabrics involving calcite sparite in a micritic groundmass. There is no verification of sponges in these cases and almost all of them can be otherwise explained; some are certainly not correctly identified. The diversity of structures seen in thin sections may be reinterpreted to include: a) meiofaunal activity; b) layered, possibly microbial (spongiostromate) accretion; c) sedimentary peloidal to clotted micrites; d) fluid escape and capture resulting in birdseye to vuggy porosities; and e) molds of siliceous sponge spicules. Without confirmation of keratose sponges in ancient carbonates, interpretations of their role in ancient carbonate systems, including facies directly after mass extinctions, are unsafe, and alternative explanations for such structures should be considered. This study calls for greater critical appraisal of evidence, to seek confirmation or not, of keratose sponge presence.