So – it’s official. Apple and Google are partnering up, and Siri is getting powered by Google’s Gemini models.
My honest reaction?
This is a good thing.
There’s been a lot of noise online from people who are aggressively anti-Google, and look – I get it. I’ve literally just finished moving my own data off Google and onto my Synology NAS because I don’t trust them with my data. More broadly, I don’t really trust any company whose entire business model revolves around collecting your data and giving you a “free” service in return.
It’s also why I don’t have Facebook, Twitter/X, or Instagram. Once a company proves itself untrustworthy, I do what I can to get my data off their platform. Simple.
That said, I think a lot of people are missing a pretty important distinction here.
Meta, X, and Google are data companies. Their products exist because of data collection. Apple, on the other hand, has always been a hardware company first. Yes, they sell services – but their margins, incentives, and long-term strategy are fundamentally different.
Apple has spent years banging the privacy drum, and unlike most companies, they’ve actually built systems to back it up. That’s also probably why they’ve struggled to build a truly competitive AI platform so far. Training large-scale models requires enormous volumes of user data – something Apple has deliberately avoided hoovering up. Combine that with the internal AI drama we’ve all heard about, and it starts to make sense why Siri has lagged behind for… well, basically forever.
From everything I’ve read, the Gemini models won’t be running on Google’s servers. They’ll be running on Apple’s own Cloud Compute infrastructure, with anonymised requests and no long-term storage of user data. Apple won’t know who you are, and Google won’t see your requests at all.
Yet somehow, a lot of people have jumped straight to “Google now has all your Siri data.”
That simply doesn’t line up with how this is being implemented – or with Apple’s entire privacy positioning.
Now, in terms of optics? Yeah… it is a bit mad that a trillion-dollar company fumbled AI this hard. But there’s an old saying that fits perfectly here:
If you can’t beat them, join them.
Apple hasn’t “joined” Google in the way some people are framing it. They’ve picked the best AI engine available right now, run it on their own infrastructure, and bought themselves time. Time to ship something competitive today, while continuing to work on their own models in the background.
They’ve done this before. Before Apple Maps existed, Google Maps was the default on iPhone. When Apple Maps launched, it was… let’s be honest – pretty whack. But over time, it improved massively. If Apple learns from that mistake, they’ll only ship their own LLM when it’s actually ready, not when it’s half-baked.
So am I sceptical? Not really.
Privacy is still Apple’s biggest selling point, and I don’t see them torching that reputation lightly – especially not with products like HomePods, wearables, and whatever “Apple Glasses” eventually become. None of that works without a genuinely capable AI engine behind it.
And let’s be real for a second…
I’m just excited to finally have a Siri that isn’t shit.
I’ve been an Apple fan boy for years. There are loads of upsides to the ecosystem – but Siri has always been the weak link. If this partnership fixes that while Apple keeps its privacy promises intact?
Yeah. I’m here for it.
