I don’t know the particulars of that genus of plants, but if the photos and any other evidence provided don’t allow distinguishing between two or more species, I would add a disagreeing ID at genus level.
If the observer comes back and uploads more photos, or says “well I measured the style/leaf width/looked at the hairs on the such-and-such and it shows that it’s Eschscholzia sp. X” that would add more nuance as far as appropriate etiquette for existing and subsequent IDs.
The ladybird/insect text on the getting started page was written before explicit disagreements were built into iNat, so that section should probably be updated. That said, it’s still not really part of iNat culture to go around to observations that are already IDed to species and say “I know this is an insect, but I can’t confirm it’s a seven-spotted ladybird” because that would lead to a bunch of annoying notifications in the current system. i.e. you should really only add an ID if it’s constructive or confirmatory.
Chiming in that you agree something might be better IDed only to genus sounds constructive to me. There will always be some nuance and exceptions to the “rules”.