i don’t see where you have a lot of old observations that are at family or higher which need Ids. where you do have a few, you either have things that may be hard to identify to species (ex. invertebrates, observations with distant photos, observations of species which have 1 or more very similar species, etc.) or which are casual observations.
the plants that i see look identifiable, and the computer vision seems to be giving reasonably good suggestions in most cases. for example:
here’s someething that can give you a general idea of how likely different “iconic taxa” will ever reach research grade (which also gives you an idea of how quickly low-level identifications will come through): https://jumear.github.io/stirfry/iNat_obs_counts_by_iconic_taxa