Thread

1. cainxinth · 1790521682 · focus · HN ↗

   These are all accurate and bad, but I have to wonder how many of these issues are just our current “ai can’t do hands” that will eventually be solved.

   1. gpm · 1790522001 · focus · HN ↗

      All the alignment issues seem likely to be solved as soon as LLMs can actually do vision well IMO.

      1. minimaxir · 1790522876 · focus · HN ↗

         All modern multimodal models can do vision sufficiently well for web design, with the exception of respecting negative space and seeing poor padding/margins on text. The issue is that prompts are often egregiously underspecified so the design -> vision loop doesn't know how to refine.

         1. gpm · 1790547242 · focus · HN ↗

            There shouldn't need to be any prompting to result in correct alignment - like icons being centred. Under-specified prompts can justify the stylistic complaints where all the AI sites favour the same few designs, but not the simply erroneous positioning of things.

            I couldn't swear the issue is the vision and not a failure to choose to correct the issues after seeing them - but from interacting with them in other areas I suspect the former not the latter.

Unofficial Hacker News client; not affiliated with Y Combinator.