# Subspecies: Views are changing, what about you? **URL:** \ **Category:** Nature Talk\ **Created:** [September 18, 2026, 10:19pm UTC](https://forum.inaturalist.org/t/subspecies-views-are-changing-what-about-you/85592 "2026-09-18T22:19:25Z")\ **Posts on this page:** 1\ **Showing post:** 13
**Author:** ![upupa-epops](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.inaturalist.org/upupa-epops/32/1972_2.png) [@upupa-epops](https://forum.inaturalist.org/u/upupa-epops)\ **Post date:** [September 22, 2026, 9:00pm UTC](https://forum.inaturalist.org/t/subspecies-views-are-changing-what-about-you/85592/13 "2026-09-22T21:00:15Z")
> [@cs16-levi](#): > > I would appreciate improving support for subspecies within the CV. The Suggestions tab will currently only provide photos and distribution maps for species observed within a particular country/state/province/county and it would be great to be able to see these at a subspecies level. Some birds are clearly different, but have almost no observations at the subtaxon level, people can’t identify them and help with research if they don’t know they exist. Some comments from staff on the subject of hybrids, which I think would have a lot of functional overlap with subspecies with respect to CV capabilities: > [@Make computer vision include hybrid taxa on an opt-in basis](https://forum.inaturalist.org/t/make-computer-vision-include-hybrid-taxa-on-an-opt-in-basis/60160/20): > > > Our CV models are trained to recognize discrete, mutually exclusive, distinct taxa. Given a photo, there should be one right answer as to what discrete taxon it belongs to. Hybrid taxa, while being potentially useful taxonomic entities, make it hard for our CV models to visually distinguish hybrid taxa from their hybridized origins, and to confidently recommend any of these taxa in any scenario given their visual overlap. > [@Force computer vision to back off on the specificity of suggested IDs in regions with cryptic or hard-to-identify species](https://forum.inaturalist.org/t/force-computer-vision-to-back-off-on-the-specificity-of-suggested-ids-in-regions-with-cryptic-or-hard-to-identify-species/70021/99): > > > Overlapping visual clades … heavily confused the model and degraded its overall accuracy The developers [are working](https://www.inaturalist.org/blog/133388-our-product-goals-for-july-december-2026) on improving these issues, which would also help it be functional for subspecies. However it does tend to seem like iNat’s official priorities are more towards giving simple species names to laypeople rather than getting into the intricacies of subspecies and species complexes. --- _[View the full topic](https://forum.inaturalist.org/t/subspecies-views-are-changing-what-about-you/85592)._