I might be completely wrong, but by making “pro-privacy bets”, they shot themselves in the foot. Average customer ended up really not caring about the benefits that we try to promote, which hinders the development and features in some sense. Also, it’s still a pain for them to integrate AI in their second biggest market — China.
Apple doesn't care about privacy in China. Even iCloud data on China was held by a Chinese custodian. The pain they experienced to integrate AI in China was because of the rule that Apple must choose a Chinese partner to integrate with. So they cannot choose OpenAI like they did in the rest of the world. They also couldn't choose Google Gemini or Anthropic obviously. They evaluated at least Alibaba's Qwen and DeepSeek and maybe some more.
Maybe go all in on local on device low latency LLMs? They make their own hardware so they could probably do it and do it well with enough focus and investment.
8GB of RAM is not perfect for this. The M chips are only known for AI because unified memory allows you to get huge amounts of "VRAM", but if you're not putting absurd amounts of RAM on the phone (and they're not going to) then that advantage largely goes away. It's not like the NPU is some world-beating AI thing.
That seems like the opposite of what they were suggesting? Unless by “edge compute” you meant user’s devices, but I assume you intended the usual meaning of CDN edges.
I care about privacy but Siri is so immensely frustrating and terrible that at this point I'd rather just prefer a competent integration with an AI that actually works.
I care about privacy too, that’s one of the reasons why I’m within Apple garden. But it definitely adds some hurdles when you want to do stuff like “literally take all of my data from every possible place and do stuff with it”.
