The primary advantage of dedicated hardware accelerators for the past ~15 years been power efficiency, rather than performance. Not much of a problem on a desktop, but very noticeable on mobile devices.

8K decoding needs HW acceleration, unless having 8-16 cores with AVX-512.

Yes, but that's way out of budget for most consumers.

Actually not really, Lunar Lake chips with Xe2 graphics have VVC decoding in hardware which can do 8K.

https://en.wikipedia.org/wiki/Versatile_Video_Coding

They were likely saying that 8K displays are way out of budget for most consumers.

I could also imagine the decoded output being pluggable into Vulkan-based filters (mpv uses https://code.videolan.org/videolan/libplacebo) without slow device->host copying.