Half the time I don't even use native rendering, I use supersampling. I have a 1440p monitor, and I often play games at 4K.
I have an RX 7800 XT, and I play a lot of older games that don't really need a GPU that powerful in order to give good performance at native 1440p. Compared to the game's built-in MSAA implementation, 4K scaled down to 1440p often looks just as good IMO, and runs better.
If I had an Nvidia GPU, and played games that support DLSS, I might use it, but I don't, so no.
I really hate jaggies and other aliasing artefacts. I also hate blur and temporal artefacts. I want a high-detail image, which looks as close to reality as possible.
DLAA is (at least in theory, assuming a good implementation, though that also applies to any other AA method) the gold standard of image quality and performance available at the moment, but none of the cinematic games I regularly play support it, so I stick to supersampling.