I believe wholeheartedly that AVIF's approach is significantly better UX.
What exactly is AVIF doing in this case that I can't do with two <img> tags at a smaller size? I can easily create a thumbnail that beats the AVIF quality in less than the 5KB delta between the static and progressive version in Squoosh.
why? in jpeg xl i get an image quicker and i can see it getting better, so i know it's still loading while the avif one looks stuck for the majority of the time, which makes it look already done loading and just low quality to me.