The terms the two of you are looking for in your discussion here are Pareto efficiency, and Kaldor-Hicks efficiency.
Your point is that not everyone is better off, so destruction is not a pareto improvement. Their point is that on net, value is increased, so in theory the ones who are worse off could be made whole through a wealth transfer, and there'd still be a surplus, a Kaldor-Hicks improvement.
You raise an important distinction, but one thing I've always found to be troublesome with K-H is that I've never seen a discussion of whether the winners would have undertaken their profitable behavior but-for the high level of profit. Perhaps if the gentrifiers (as just one example) faced a gentrification tax to fund the lives of people displaced they would not have considered the deal good enough to gentrify in the first place.
I'm not saying this is necessarily true, but it seems like a major problem with K-H as a concept that I personally don't see discussed (I imagine it is discussed in academic journals somewhere).
Your point is that not everyone is better off, so destruction is not a pareto improvement. Their point is that on net, value is increased, so in theory the ones who are worse off could be made whole through a wealth transfer, and there'd still be a surplus, a Kaldor-Hicks improvement.