The reason for Apple to do this will be exactly why Microsoft is happy for OpenAI to provide the Models. If anything goes wrong it's not Apple or Microsoft's fault. It is the SaaS provider who is to blame.
For apple this will mean Google's stock takes the hit while theirs remain unaffected. Unfortunately for OpenAI/MS, OpenAI is not public and so everyone is using MS as a proxy for trading on OAI. So if something goes wrong in the Bing/CoPilot world, MS stock will take the hit.
Given all the regulatory scrutiny on the Google-Apple search deal, I believe we'll get an option to choose the 'AI Backend' similar to how EU users get to choose Search Engine.
In effect a lot more people are aware of Google logo than OpenAI so that's what most common people will choose, while maybe most tech-savy folks will end up choosing OpenAI.
This will be Apple's way to hedge against any further regulation option and they get to market iPhone as a AI aggregator similar to the AppStore.
It either indicates Apple isn't sure it can catch up or that they simultaneously realize they need to provide these AI features but don't think AI as a marketing component will have the staying power to make the investment worth it.
I can't say I disagree very much, as the AI hype is quite overblown.
That would also explain why they didn't pick something like Mistral or Stability. Newcomer with no strong legal dept can easily be tanked if people catch on that training infringes and legal winds change. Google is less likely to fail catastrophically when it comes to that.
Is Apple making profit out of generating and selling derivative works from music and movies without compensating the original authors? If not, I'm not sure how it relates to the current breed of generative AI.
Mine being that large language and diffusion models are trained on "whatever the heck is there, available for scraping on the broad internet, indiscriminately".
I don't think Apple's money or it being in bed with the Majors solves either side of "acknowledging authorship when none is provided/known", "obtaining formal consent for commercial use" or "fairly compensating a creators base" when it is larger than all humans on earth.
I agree that this is very different from traditional content agreements. Apple is a very conservative company. They may not want to be trailblazers when it comes to working out how compensation of creators is going to work in an AI context.
What appears to be happening right now is that Google, OpenAI, et.al. are negotiating individual agreements with large publishers if and when threatened with legal action.
I'm not sure this will be sufficient. There will be calls to introduce some sort of AI tax or other forms of collective compensation schemes to funnel money to independent creators, small businesses, museums, etc.
This is all very much up in the air and it will take many years if not decades to settle.
Apple is doing it, but doubt they'll use something in-house that's still very much a work in progress when they can actually get paid by Google to use Gemini this year.
