They won't, its literally part of their sales funnel. They've specifically engineered a bad experience for anyone outside the ecosystem by making it all of their friend's problem too. Its very important for their stock price that text messages sent by non apple products are just slightly more difficult to read.

That wouldn't let them say “AI”.

The problem is using a language model to assess images.

Probably 80% of "LLM's are below expectation" complaints (from the general population) involves some form of image analyses.

Image tokenization is hard because unlike language tokenization, where every token is extremely dense with meaning, image tokens tends to be meaningless or irrelevant but are processed all the same.

Give an SOTA LLM a picture of toothpicks and ask it to move one to make a square, and it will probably struggle and fumble it. But give a mid-size LLM from 2 years ago the same problem in verbal form, and it will nail it almost every time.

That takeaway is, do everything you can to avoid having the LLM need to rely on images for the answer.

Most of those videos are chatGPT voice mode, which still used gpt 4o last time I checked. it is far from SOTA.

Like coding, creating images or text, maybe the alternative of doing it yourself is too easy or enjoyable for you. Don't expect that will be true for everyone.

Maybe on a smartphone, but usually not on a computer. Keyboards are pretty good.

The other thing is that if you're around others, voice input means you have no privacy. Even if you're not doing anything particularly private, it's a bit awkward and potentially embarrassing. If you use touch input in conjunction with a screen reader, you can be more like a "normal" user in that what you're doing is just between you and your phone.

Audio input is far more commonly used by people with mobility difficulties. Imagine that your hands shake a lot or that you don't have limbs. That makes using a keyboard and mouse difficult or impossible and voice input can help. Blind users generally use keyboards for input, the typical ones you find on a PC but also sometimes the keys on their Braille display.