There is a list of psychoactive plants on the wiki, it's not very long. In principle, you could make a list of all known plants, which would be very long.
As a percentage, how many plants in the second list should be on the first list, but aren't because their psychoactive contents have not been discovered?
An answer of 100% would mean that you think that all plants are psychoactive. An answer of 0% would mean that you think that there are no psychoactive plants, except for the ones on the wiki.
I'm considering trying to automate as much of the extraction process as possible (mostly just because I like robots). If it becomes easy and reliable enough, I might just start running extractions on random samples that I find in the wild. Of course there would need to be a whole lot of verification work if I ever found something, but I think it would be a positive contribution--which I'd like to make.