Apple Used Google Tensor Chips to Develop Apple Intelligence

[MacRumors]

Apple used Tensor Processing Units (TPUs) developed by Google instead of Nvidia's widely-used graphics processing units (GPUs) to construct two critical components of Apple Intelligence.

The decision is detailed in a new research paper published by Apple that highlights its reliance on Google's cloud hardware (via CNBC). The paper reveals that Apple utilized 2,048 of Google's TPUv5p chips to build AI models and 8,192 TPUv4 processors for server AI models. The research paper does not mention Nvidia explicitly, but the absence of any reference to Nvidia's hardware in the description of Apple's AI infrastructure is telling and this omission suggests a deliberate choice to favor Google's technology.

The decision is noteworthy given Nvidia's dominance in the AI processor market and since Apple very rarely discloses its hardware choices for development purposes. Nvidia's GPUs are highly sought after for AI applications due to their performance and efficiency. Unlike Nvidia, which sells its chips and systems as standalone products, Google provides access to its TPUs through cloud services. Customers using Google's TPUs have to develop their software within Google's ecosystem, which offers integrated tools and services to streamline the development and deployment of AI models.

In the paper, Apple's engineers explain that the TPUs allowed them to train large, sophisticated AI models efficiently. They describe how Google's TPUs are organized into large clusters, enabling the processing power necessary for training Apple's AI models. Apple has announced plans to invest over $5 billion in AI server enhancements over the next two years, which should bolster its AI capabilities and reduce its dependence on external hardware providers.

In addition to detailing its use of Google's TPUs, the paper addresses ethical considerations in AI development. Apple emphasized its adherence to responsible data practices, claiming that no private user data was used in training its AI models. The company relied on a mix of publicly available, licensed, and open-sourced datasets for training purposes. Apple added that its training data set, which includes publicly available web data and licensed content, was curated to protect user privacy.

Article Link: Apple Used Google Tensor Chips to Develop Apple Intelligence

 

[GrayFlannel]

Apple scrambling to slam it together.

Here’s two articles on it too, from yesterday.

Apple admits to using Google Tensor hardware to train Apple Intelligence

New artificial intelligence research published by Apple reveals that Apple has been using Google hardware to build the early foundations of Apple Intelligence.

appleinsider.com

https://www.patentlyapple.com/2024/07/apple-confirms-that-its-ai-models-were-trained-on-googles-custom-tensor-processing-unit-tpu.html

 

[nono_thank you]

Apple scrambling to slam it together.

If you can't beat 'em....

 

[nt5672]

So while Apple may not have used user attributed data, Google still knows exactly what Apple is doing in AI, exactly how their models perform, and can compare their own AI with Apple's AI since they have both.

Sounds like a win-win for Google, Apple not so much. I wonder how much intellectual property Apple is giving up just to say "We can do that too!"

This sounds like a pretty big failure at the executive level.

 

[Chungry]

Securing Nvidia chips takes time and money- Apple only had one so they settled for what they could get. I’m concerned about the precedence this sets, because it caught them flat-footed and they’re in damage control mode. Compromising on the foundational piece instead of delaying like they say they do- going with best of breed, and not being first but being best. This is a move exuding an utter lack of confidence. It’s got a shiny veneer though! Leaves me concerned for Apple Intelligence. Let’s hope that doesn’t become a running gag like the mention of Siri.

 

[vegetassj4]

Hey Siri: Don't be evil

Siri: Sorry, I didn't quite get that.

 

[MacHelp_75]

There is a history between Apple and Nvidia. Lot's of bad blood over GPU's led them to AMD prior to Apple Silicon. If senior leaders are still around who lived through that it may have affected the choice. Speculation obviously, but one supported by history.

 

[Jarman74]

So not only was Apple reliant on Google services, but it was also comically lagging behind Nvidia’s hardware offerings. It makes you wonder what outdated tech Siri is still running on.

 

[Jarutais]

So.. are they acknowledging that the Apple Silicon Neural Engine is crap? I thought they always marketed it as the best for machine learning

 

[bLackjackj]

So not only was Apple reliant on Google services

Apple IS already reliant on Google services. Where do you think icloud resides?? Of course they are going to chose a partner they are already doing business with. Thats called business sense.

 

[SmugMaverick]

Absolutely comical from Apple, you couldn't make this up.

Google tensor gets a load of stick from tech "journalists" but its still the best for AI.

Dumping iPhone 15 pro for Pixel 9 Pro on the 13th, this is just embarrassing from Apple.

 

[Yammabot]

Apple was already in bed with google but this is a new leap forward that deepens the privacy stance of the company. How can you promise to maintain privacy while usi googles servers? That’s idiotic! Are they giving google all our data? Even if they remove the identifiers it’s still finger-printable.

 

[lkrupp]

There is a history between Apple and Nvidia. Lot's of bad blood over GPU's led them to AMD prior to Apple Silicon. If senior leaders are still around who lived through that it may have affected the choice. Speculation obviously, but one supported by history.

I was going to say the exact same thing. Thanks

 

[SmugMaverick]

I've said it before but Google was really on the right track picking AI on Tensor over raw power like Apple, no apps require the raw power Apple touts all the time and tech dweebs bang on about in speed tests.

Tensor was MADE for AI, if the rumoured gemini nano update is true on 13th then Apple " intelligence " is going to be so far behind by the time it's fully released in 2025.

I can't believe Google has a vision for AI while Apple wasted time on a useless headset vanity project for dim Tim.

 

[iBluetooth]

I was going to say the exact same thing. Thanks

For those that don't know the past history of Nvidia on Apples black list. A lot of MacBook Pro computers had a failed Nvidia discreet graphics, resulting in a recall by Apple. Then Nvidia denied any claim and left Apple paying for all the repairs and replacement parts, even paying Nvidia for new parts. This was before Apple came this wealthy.

 

[krspkbl]

Apple is choosing to hold a grudge over picking the better option.

Who uses Google for AI stuff lol

Apple should have been building AI infrastructure over the last 5-10 years but of course they've been caught out and are lagging behind. It's a shame that they need to rely on others. Why exactly is Apple worth 3+ trillion and what are they doing with their money? Financing TV/Movies nobody wants to watch, developing hi tech ski goggles nobody wants, and stuffing the pockets of share holders.

 

[JPack]

This disclosure feels like a way out when Apple Intelligence doesn’t perform as expected.

“It’s because we used Google Tensor chips.”

Either that or Google paid them to use Tensor. Apple rarely discloses the hardware they use to develop stuff. There was no reason to do so here, especially given the cloud based computing and the privacy stance they have towards that.

 

[Contact_Feanor]

So while Apple may not have used user attributed data, Google still knows exactly what Apple is doing in AI, exactly how their models perform, and can compare their own AI with Apple's AI since they have both.

Sounds like a win-win for Google, Apple not so much. I wonder how much intellectual property Apple is giving up just to say "We can do that too!"

This sounds like a pretty big failure at the executive level.

This isn't exactly true. Cloud based infrastructure usually has pretty good checks in place to protect customer privacy exactly because of this sort of thing. Microsoft can't/doesn't look at everything that's on azure, amazon doesn't look at everything on their web services exactly for this reason.

 

[krspkbl]

This disclosure feels like a way out when Apple Intelligence doesn’t perform as expected.

“It’s because we used Google Tensor chips.”

Maybe they'll eventually go Nvidia once the old dinosaurs are retired... or maybe they want to eventually make their own AI chips and make them look better without doing much work. Don't set the bar too high lol.

 

[bradman83]

Hell hath no fury like Apple leadership's grudge against Nvidia.

 

[WB2Colorado]

Apple and Nvidia haven’t been on good terms for many years now so it isn’t exactly a shock that they wouldn’t use them for AI stuff

 

[krspkbl]

Hell hath no fury like Apple leadership's grudge against Nvidia.

It'll only have deepend because Nvidia are no longer just a GPU company. They are the leader when it comes to AI hardware. Nvidia have been playing the AI game much longer than the likes of Microsoft or Google. Their consumer GPUs have had AI tensor cores for about 4 years now. That's why they have grown significantly recently. They saw the future vision. Everyone else jumped on the bandwagon and Apple is last to get onboard.

 

[Chungry]

There is a history between Apple and Nvidia. Lot's of bad blood over GPU's led them to AMD prior to Apple Silicon. If senior leaders are still around who lived through that it may have affected the choice. Speculation obviously, but one supported by history.

That bad blood runs deep, I was around for that and remember the G4 fiasco. It’s silly that Apple would seemingly cut off its nose to spite its face and, in this instance, offer an also-ran solution for what at this point can only be described as pure spite and pettiness.

So not only was Apple reliant on Google services, but it was also comically lagging behind Nvidia’s hardware offerings. It makes you wonder what outdated tech Siri is still running on.

Using Siri removes all wonder. It’s comically bad outside of a time or turning off a light. Amazon had this licked ages ago and they suck compared to the modern assistants.

So.. are they acknowledging that the Apple Silicon Neural Engine is crap? I thought they always marketed it as the best for machine learning

It’s a curious self-own but that’s my takeaway too.

Absolutely comical from Apple, you couldn't make this up.

Google tensor gets a load of stick from tech "journalists" but its still the best for AI.

Dumping iPhone 15 pro for Pixel 9 Pro on the 13th, this is just embarrassing from Apple.

I’ll be moving from my 13 Pro Max to a Pixel 10 next year since nothing is really keeping me in the Apple ecosystem from an app stickiness standpoint. I’ll still have a 15 Pro Max for work to compare and contrast but Apple is so staid and boring with their phone designs and restrictions - it’s opinionated design which isn’t for everyone. Funny, it was superior for so long but Android really blew past them in the last two years. I think the Pixel 9 looks unique and engaging. Who knows how long until we see a compelling design in a new iPhone. Which is a shame because new-looking phones are absolutely a major mover for products. I’d say this is partially to blame for their slump. But hey, the iOS 18 stuff is a glass of ice water for folks in hell. Control Center still isn’t as customizable as I’d like it to be, the Home Screen stuff is mostly catch-up for feature parity’s sake.

On a nitpick note: Can I remove icon labels without going into large icon mode Apple? Just why is that a restriction?

 

[GrayFlannel]

Securing Nvidia chips takes time and money- Apple only had one so they settled for what they could get. I’m concerned about the precedence this sets, because it caught them flat-footed and they’re in damage control mode. Compromising on the foundational piece instead of delaying like they say they do- going with best of breed, and not being first but being best. This is a move exuding an utter lack of confidence. It’s got a shiny veneer though! Leaves me concerned for Apple Intelligence. Let’s hope that doesn’t become a running gag like the mention of Siri.

They were more than caught flat-footed. The board and shareholders had to put a boot in their butt.

 

[JPack]

Apple and Nvidia haven’t been on good terms for many years now so it isn’t exactly a shock that they wouldn’t use them for AI stuff

Apple has very bad relations with Qualcomm, but they ended up going back to them for best performance.

In this case, it’s clear Nvidia has the best hardware for developing AI.