A speech or LLM based device is only more simple if it works perfectly. As it stands today, it’s far from perfect. When it makes a mistake, if there nothing to fall back on, I would think that would be very frustrating. I run into these types of issues on a daily basis with current LLMs.
I would liken it to a non-responsive touch screen. The magic of the modern smartphone evaporates if it stops responding to touch.
I missed the early iPhone as well. I actually setup my Home Screen on my phone to be the same app setup as the original iPhone when iOS 26 released. I still have other apps in the App Library if/when I need them, but I have my basic setup how things were in 2007… At least for now. I’ve actually liked it quite a bit so far.
I’m curious if Siri with Apple Intelligence will live up to some of what they showed in the ads last year, but at this point I need to experience it working with 0 issues for an extended period of time before I even start to think about an AI-first device, and especially for an AI-only device. I haven’t seen AI from anyone that can perform at that level. Much like full-self-driving, I feel like this is going to be something that is perpetually 5 years away.
Comments
A speech or LLM based device is only more simple if it works perfectly. As it stands today, it’s far from perfect. When it makes a mistake, if there nothing to fall back on, I would think that would be very frustrating. I run into these types of issues on a daily basis with current LLMs.
I would liken it to a non-responsive touch screen. The magic of the modern smartphone evaporates if it stops responding to touch.
I missed the early iPhone as well. I actually setup my Home Screen on my phone to be the same app setup as the original iPhone when iOS 26 released. I still have other apps in the App Library if/when I need them, but I have my basic setup how things were in 2007… At least for now. I’ve actually liked it quite a bit so far.
I’m curious if Siri with Apple Intelligence will live up to some of what they showed in the ads last year, but at this point I need to experience it working with 0 issues for an extended period of time before I even start to think about an AI-first device, and especially for an AI-only device. I haven’t seen AI from anyone that can perform at that level. Much like full-self-driving, I feel like this is going to be something that is perpetually 5 years away.
Agree that if the primary UI is voice, it will need to be the the multiple nines of success to avoid being frustrating.
Even so, I still frequently use Siri now despite it being much less successful than that.