They want tools that do things and toys that are fun. So maybe. It depends on what Apple will use it for. I enjoy being able to search my photos even though I never tagged them. That’s a useful kind of AI. I like how I can automatically select the subject of a photo so I can place it on other backgrounds also.

Doesn’t matter what customers want, it’s the shareholders.

Do the majority of users really want AI in their computers?

What this could mean is the ability to replace (or upgrade) something like Siri into a model that runs locally on your machine. This means that it wouldn’t need to route your questions/requests through someone else’s computer (the cloud). You wouldn’t even need to connect the computer to the internet and you would still be able to work with that model.

Besides, there are many companies that don’t want you to pass on their internal documents to companies like OpenAI (ChatGPT). With locally run models, there aren’t any problems with this as that data will not be uploaded anywhere.

I mean that depends? Does it actually work? Does it record all my data constantly and send it to Apple in a continuous stream, or is it fully local? will it actually provide some useful service like better search, will it be able to bootstrap projects for me? Learn my workflow patterns and assist by preparing stuff preemptively, or make intelligent suggestions on how to improve efficiency? Can it attempt to organize my awful, trash filing system of pictures, movies, half-started projects into something less painful to look at? Can it help my computer get out of my way when I don’t feel like being all computery (which I love but sometimes it shouldn’t be the focus)? Can it do these things well and without leaking them to some cloud?

Because that’s what we’ve been trying to make computers do since the PC first came along and I’d fuckin pay for that.

EDIT: to more directly answer your question, no one wants SaaS AIs by the FANG because they’re labor-replacement machines aimed at the ultimate grift; extracting capital out of the capitalists at the expense of the everyday workers. But that doesn’t mean that every AI is a speculative language model trained on the stolen data of all humanity to be used as a tool for class warfare; if they can create a neural net that can focus on other things besides language (or hell? Even language) that runs locally- if they can eat Microsoft’s lunch and create an actual co-pilot; they’d charge a premium as they always do but if it works well enough, it might be worth it.

This always happens when something new and novel has “potential”. VC money has been funding loss-leaders for two decades and they wanna cash in on the next gold rush. Just like blockchain, expect to see this beaten to death and shoehorned into places it really has no real use. They’ll be a few really solid things that are found for it to do though, and it will excel in those places. Then we’ll all laugh about “remember when they thought LLMs were the next big thing? What a bubble that turned out to be, like pets.com all over again”

Depends on your work, what you’re trying to do, and how you use it.

As a developer I run my own local version of Dolphin Mixtral 8x7B (LLM) and it’s great at speeding up my productivity. I’m not asking for it to do everything all at once but usually just small snippets here and there to see if there’s a better or more efficient way.

I, for one, am looking forward to hardware improvements that can help us run larger models, so news like this is very welcome.

But you are correct, a large number of companies misunderstand how to use this technology when they should really be treating it like someone at an intern level.

It’s great to give small and simple (especially repetitive) tasks, but you’ll still need to verify everything.

Just give it a couple of years for the hype/boom/bust cycle to complete, then it’ll settle down and people will start using the tech appropriately.