I haven't heard of, and can't think of, a solution to this problem that doesn't introduce much larger problems. Big picture, I don't think the problem of distinguishing humans from bots online is a big enough deal that we should take hits in other areas to solve it.

The least-harm solution, as far as I can see right now, is to just accept that the internet cannot be made trustworthy in this way. The only way to know for sure the nature of who you're dealing with will be to deal with them in person. Much like it has always been.