Apple recently announced its next crack at integrating “AI” into its operating systems, this time opting to simply whitelabel Google’s Gemini “AI” tools instead of developing its own LLM technology. Called “Siri AI”, Apple also stated it’s not coming to the EU, and the company stated that’s because the EU’s basic consumer protection legislation would give other “AI” tools “unprecedented access” to user data on users’ devices. The company made a big stink about this in the press.
As anyone with basic pattern recognition skills already knew, this was a blatant, baldfaced lie. What really happened is that Apple asked the EU for an 18-month long exemption from the EU’s consumer protection and privacy legislation during which it would not have to comply with any legal privacy and interoperability requirements – just so it could roll out Siri “AI” before anyone else could offer a competing product for Apple users.
Obviously, the EU wasn’t going to grant such an exemption.
“The decision not to roll out Siri AI in the EU is Apple’s and Apple’s only,” spokesperson Thomas Regnier told reporters in Brussels, saying there was nothing in the Digital Markets Act to stop the company from introducing new products in the EU.
“Apple was simply unable to develop interoperability solutions that meet essential EU privacy and security standards,” Regnier said.
“Instead of trying to find a suitable compliance solution, Apple simply made a request to the European Commission to be exempted from their interoperability obligations under the DMA – and this for at least 18 months. That’s not an option,” Regnier said.
↫ Inti Landauro and Foo Yun Chee at Reuters
So what’s really going on here is that Apple wants to offer a set of whitelabeled Google Gemini tools on iOS and macOS in the EU, but because Apple is classified as a gatekeeper, it is legally obligated to offer interoperability options for competing “AI” tools. These options in turn need to adhere to the EU’s strict privacy regulations, so that competing “AI” tools can offer the same level of privacy that Apple’s own whitelabeled Google Gemini tools claim to offer.
Apple didn’t want to offer these privacy-respecting interoperability options as required by law, so instead of following the law in the countries it wants to operate in, Apple asked to be placed above the law for at least 18 months, basically giving Siri “AI” a massive head-start over possible competitors so that it could entrench itself in the userbase. The EU saw right through Apple’s nonsense, and now called them out on their bullshit. Perhaps Apple has gotten so used to openly bribing Trump that they forgot other parts of the world don’t work that way.
Whenever Apple and its PR attack dogs say anything about the EU, you can be assured they are lying. They have proven time and time again to basically never speak a single word of truth when it comes to its dealings in the EU. It’s almost pathological at this point, and what makes it doubly interesting is that Apple will not launch Siri “AI” in China either, for the very same regulatory reasons – yet all China got was a single footnote in a press release.
I wonder why.
Stealing a comment from Reddit …
The issue is basically that Siri AI is given access to Spotlight, email, photos, messages, contacts, location etc. Apple doesn’t want some random ass app from the App Store to have similar access. They could bundle in a bunch of consent screens, but then (1) Siri would have to constantly bombard you with cancel or allow dialogs, (2) some fly-by-night app would probably cause another fappening-style disaster or worse anyway because people click through those.
To put it another way, the EU rules would require Apple to offer Grok, DeepSeek, and other dubious AI chat bots unlimited access to your data and Apple said no.
Once again, the DMA is pretty good for EU businesses like Spotify, kind of bad for Apple, and really bad for end users.
WorknMan,
IMHO everyone should have the right to choose the AI they want (including none at all).
I feel the EU is right, apple SHOULD NOT be giving it’s own AI exclusive access to user data. If a user wants to use DeepSeek, Grok, Gemini, or heck even a FOSS model running privately at home, that choice needs to be made by the user. Of course apple should warn users about privacy risks, but apple doesn’t deserve to deprive consenting users of choice to reinforce apple’s monopoly.
Alternate viewpoint, Apple doesn’t want to force Siri AI to have all the same permissions alerts/prompts as everything else would need, and so since they don’t want to privilege their own system over everyone else, they’re taking their ball and going home. And not designing a safe/secure way of having any of these AI assistant apps to interact/use your phone.
The day I won’t have to use the official client to connect to whatsapp//iMessage is the day DMA will have won.
> To put it another way, the EU rules would require Apple to offer Grok, DeepSeek, and other dubious AI chat bots unlimited access to your data and Apple said no.
Yep, and Apple is wrong on this.
The end user is in charge of his device, having it any other way is rent-seeking.
Responsibility should also move : if the user installs an evil AI, that is not Apple’s responsibility.
Serafean,
I agree with both these points. Apple should not be usurping owner rights, but also apple aren’t responsible for 3rd party services. The real reason apple are fighting this lines up with what Thom is saying, apple are anti-competition to their core and they built an empire upon blocking competition. Apple are fighting the EU because the EU is defending an owner’s right to chose competing services, which means apple looses the ability to leverage iphone dominance to promote their own services.
Having a choice of appstores is really bad for end users?
Having a single cable work for everything after decades of cable and port hell is really bad for end users?
Putting an end to kill switches implemented in games they bought is really bad for end users?
This end users folk must be really masochist fellas.
Roll out a feature in EU? Anticompetitive.
Don’t roll out a feature in the EU? Believe it or not, anticompetitive.
DefineDecision,
Well, it’s not really about the feature per say, rather it’s about apple’s use of vendor locking. The EU wouldn’t take issue if owners had the right to choose competing alternatives, that is what’s inherently anti-competitive.
Was just listening to a podcasts of the journalists that were present at Apple’s separate, after-keynote AI tech talk with all the Apple big names involved and what was made very clear by Apple, as relayed by direct quotes from the journalists, is that no, this is not a white label of Google’s AI. These are Apple models either run on device, on Apple cloud servers, or in the most challenging cases, still Apple models but run on Google’s larger hardware with lots of privacy protections around them.
Specifically the Upgrade podcast at around 47 minutes: “…but it’s also not just a white label of Gemini. It’s just not.” There’s a good amount of talk about how the models are Apple models before and after that, as well.
So yeah, I think the base premise of this story is… inaccurate.
And then we get to the stuff that WorknMan is saying. While some of the EU rules around competitive markets are good and necessary, a lot are definitely not. You want to make a common data cable connector like USB-C? Eh… sure. But dictating to companies that there can be no competitive advantage and that any technical advance must be shared and swapped out with anyone and everyone no matter the cost or privacy concerns. No thank you. I use an iPhone specifically to stay away from Google’s often “oops, we were scooping up all your user, ad, wifi, and location data even though you turned those features off” style of doing business. Allowing a different AI provider direct, unrestricted access to everything on an iPhone is almost certainly going to be a huge negative for most users.
Doesn’t mean I think Apple is right on everything. Hell, they basically stuck their heads in the sand about app stores and alternate payment processors for far too many years and various countries around the world have rightly called them out on it and forced them into doing something closer to the right thing. (Not that Apple is doing the right thing there even now, they mostly did aggressive, malicious compliance… but that’s a conversation for another day…)
Ragashingo,
Respectfully, I think the premise here is inaccurate. The EU isn’t demanding apple give everyone including google permission to your private data. The EU is demanding apple respect the owner’s choices, meaning the choice is yours to make, not apple’s.
Totally agree.
Thom, for the second time, this is factually wrong. It is simply untrue that Siri AI will not be coming to the EU. It WILL be available on Mac OS even in the EU. It is only excluded from iOS and iPad OS. See https://www.apple.com/uk/newsroom/2026/06/due-to-dma-siri-ai-delayed-in-eu-for-ios-27-and-ipados-27/
At the very least, a site called OSnews should get its facts about the technologies that will be present in operating systems right.
The real question is what happens to user data once it leaves the device.
The main obstacle to the local deployment of high-performance LLM is the amount of RAM required to run it. iPhones and iPads are constrained by design choices (limited RAM, lack of active heat dissipation), making cloud storage unavoidable for larger models.
However, once data is transmitted to third-party servers, there is no legally enforceable or technically verifiable guarantee against its reuse for commercial purposes (training models, reselling aggregated data), nor against its accidental exposure.
Regarding liability in the event of a data breach: under French law (and European law under the GDPR), a company is liable as soon as a breach is detected, without the need to prove intent. The GDPR is based on a reinforced obligation of due diligence.
On the other hand, obtaining effective redress as an individual remains difficult in practice. Apple can implement these measures, but there’s no guarantee that third parties with access to its APIs will do so.
Is the responsibility with the operator using the data, or with the manufacturer who deliberately designed a system incapable of protecting it locally?