I find it mind-blowing that Gmail's algorithm doesn't lean more on users to train the spam filters...instead of top-down arbitrary keyword filters & rules that constantly result in false positives.

Apparently human users are trustworthy enough to train 4,000 lb self-driving cars to not kill humans via CAPTCHA. Why the hell can't we rely more on humans to train the spam filter?

It could solve literally every issue with email deliverability today. It's clear the current algorithms are simply not sophisticated enough to do this job properly. Hell, gmail can't even properly discern the difference between an "Update" and a "Promotion." Every dev-related newsletter I'm subscribed to (all of which I want to receive) randomly alternates between those two tabs each week.

To be honest, I highly doubt Google is using any sophisticated Machine Learning algorithms inside Gmail. It doesn't seem like a big priority for them. The space is ripe for a competitor to come in.

I would LOVE to have an email address where I was 100% confident in the spam filter, didn't hide a large portion of my emails behind disappearing tabs, and had more control over what categories my messages go to. Any alternatives out there I should be using?