This is like 75% "why?", 25% "terrible headline obscures viable and worthwhile use case for the technology in question"
The accessibility/usability version discussed here intrigues me, but it's not without problems of its own (how do you train that model, for example)
add a skeleton here at some point
3 days ago