In fact, Apple is working by itself generative AI, Giant Language Mannequin (LLM) and doable ChatGPT rival referred to as, naturally, AppleGPT. Positive, the information relies on a Bloomberg report and Apple is predictably mum on the matter however, critically, how may the Cupertino tech large not be working by itself AI?
In accordance with the Bloomberg report, Apple is basing its ultra-secret mission on a studying framework often known as Ajax, from rival and typically good friend Google.
The trouble to construct some kind of chatbot and perhaps different generative AI programs has been happening since late final 12 months however, as somebody who attended Apple’s WWDC 2023 can let you know, Apple made no point out of chatbots of any sort on the June developer’s convention.
Privateness roadblock
Apple‘s hyper-focus on consumer privateness has, as I see it, considerably hamstrung its efforts to deliver any type of LLM-based chatbot to shoppers. ChatGPT, Google Bard, and Microsoft Bing are all cloud-connected and ship queries out to distant servers for fast interpretation and response (primarily based on the LLM’s huge information of how precise people may reply underneath related circumstances).
That, after all, shouldn’t be the Apple approach. Its Apple Silicon A16 Bionic’s Neural Community is native. It does Machine Studying in your best iPhone. Sending queries with all these presumably private particulars is anathema to Apple’s privateness ideas.
And but, Apple clearly can’t afford to steer clear of the siren name of generative AI. It’s a revolution that’s consuming the tech business and the pursuits of common shoppers and companies. Even with the intense scrutiny AI improvement is underneath and the lawsuits some of it is facing, nobody believes AI improvement is out of the blue going to cease or go away.
Apple has even gone so far as, in accordance with Bloomberg, creating its personal chatbot, or AppleGPT. However that is principally a extremely restricted and inner check and apparently not one which’s ever headed to shopper desktops.
What about Siri?
The place does Siri sit in all this?
Bloomberg claims that the Ajax work has already been used to enhance Siri. Which may be so, however the one Siri enchancment we’re getting with iOS 17 (at present in public beta) is the flexibility to cease beginning every voice assistant immediate with “Hey.”
I’ve little question that Apple is difficult at work determining its place within the LLM AI sphere, but it surely’s additionally clear from the report that these are early days. There is no such thing as a overarching technique, and I doubt the existential query of whether or not or not Siri may ever host AppleGPT (or no matter it is referred to as) has been answered.
Finally, that is affirmation that Apple is simply as conscious of what is going on on round it and with opponents as ever. It can pattern and check, develop and check, scrap and develop, after which check some extra. I do not anticipate Apple to inform us something about this through the anticipated September launch of the iPhone 15. Nonetheless, by the point WWDC 2024 rolls round, Apple may be able to unveil a brand new platform. Perhaps it’s going to be AppleGPT-kit, AppleLLM-Package, and even AppleGPT.
This assumes that Apple can remedy its huge privateness query. If not, AppleGPT may stay in Skunkworks indefinitely.
Discussion about this post