One is love what you do - which assumes regardless of what you’re doing, love it.
The other, do what you love - means choose things that make you love them.
They’re completely opposite in what choice they’re telling people to make.
Both of these sayings can of course be applied to evil as well as good, and be either positive or pathological in impact depending on the situation.
I've tried this with programming. After 15 years, I'm only mildly positive about programming. That has more to do with the culture though, not the activity. I wish there was a business analyst/entrepreneurial programmer hybrid role. Spot opportunities from data, then build out a solution. Or if a solution doesn't need to be built out but it's more of a human/political thing, then do that.
I'm currently a data analyst and get some room (informally) to do just that. It's a really fun role, and I'm programming way cooler things than I ever did as a software engineer.
[1] Mentioning this detail because I don't think I could've learned this any other way. Well, parents maybe, but they didn't have this attitude.
The bad ones identify solutions that work for huge orgs and try to apply them to everyone and don’t implement them. This is your typical architecture astronaut. Or, they get bogged down in perfection and are massively hands on when they should have stopped long ago. Also your typical micromanaging vp of engineering.
But one or two good ones can be utterly enormous force multipliers.
Companies that gave me the chance a bit, it seems that I always fulfilled those expectations.
If you love what you do, you didn't necessarily get there by doing what you love.
Reminds me of: a duck is a bird but a bird isn't necessarily a duck.
I agree it's not the opposite.
While it is possible to both do what you love and love what you do, often if you do what you love you are apt to find out that you don't actually love doing it. Hence the common refrain that cautions against turning your love into a job.
There's significant skill overlap, but little passion overlap.