Just to confuse things even more it’s unclear what exactly the “observation copyright” applies to, for example here I was told it only applies to additional text you add and not the observation as a whole (especially not time, place, username, taxon, etc.) https://forum.inaturalist.org/t/licensing-problem-for-paper-using-inat-data/38480/57
With that interpretation the answer would be no, you cannot restrict use of those data as the copyright does not apply to them (just to text that 99% of observations don’t even have). But that’s just one interpretation and I don’t think any of the old threads about it had a clear conclusion - GBIF obviously thinks the copyright applies to all the data.