Apple’s AI Shopping Spree: Why Google Won Over Claude

According to AppleInsider, Apple recently signed a deal to use Google’s Gemini AI model to power upcoming Siri upgrades, confirming its own internal models aren’t yet competitive. The company was also in talks with Anthropic to use its Claude model for the same purpose, but those negotiations fell apart because Anthropic demanded “several billion dollars a year” with fees that would double annually for three years. In contrast, an unconfirmed report suggests the Google deal is worth around $1 billion per year, a fraction of the $20 billion Google already pays Apple to be the default search engine. Beyond Siri, Apple already uses custom versions of Claude on its own servers and integrates OpenAI’s ChatGPT for existing Apple Intelligence features. The company will also have to use a Chinese AI model for Siri in China to comply with local laws.

The Astronomical Fee Problem

Here’s the thing: Anthropic basically priced itself out of the game. Asking for several billion a year, with that figure ballooning each year, is a wild move for a company that’s still building its market position. It makes you wonder about their calculus. Were they overconfident? Did they not want the deal? Or did they just see Apple‘s deep pockets and decide to shoot for the moon? Compared to the reported Google fee, it’s a staggering difference. And when you consider Apple is already raking in $20 billion from Google for search, the Gemini deal almost becomes a rounding error on an existing business relationship. It’s a much cleaner, cheaper play for Apple.

Apple’s Multi-Vendor AI Buffet

So what’s Apple’s play here? It’s clear they’re not putting all their eggs in one basket. They’re creating an AI buffet. Siri’s brain might be Google-flavored soon, some backend processing is already Claude-infused, and for writing tools you can summon ChatGPT. They’ve even said users will get to pick their preferred AI model down the line. This is a classic Apple hedge. They know their own models aren’t ready for prime time, but they also don’t want to be permanently dependent on—or overpay—a single external provider. It’s a stopgap strategy. They get to launch competitive features *now* while buying precious time to develop their own secret sauce in-house.

The Eventual Endgame

Look, let’s be real. The endgame for Apple is almost certainly to cut out all these middlemen. They hate relying on other companies for core tech. Remember the Intel-to-Apple Silicon transition? This is shaping up to be the AI version of that. All these partnerships with Google, Anthropic, and OpenAI are just temporary bridges. Gurman’s report notes Apple’s record R&D spend, and you better believe a huge chunk of that is going toward building their own foundational models. The goal is to have “Apple Intelligence” powered entirely by Apple silicon running Apple software with Apple AI models. It’s the whole vertical integration dream. These deals just keep them in the race until their homegrown tech is ready to take over.

What It Means For Everyone Else

For users, this multi-source approach is a mixed bag. On one hand, you might get a more powerful Siri faster. On the other, it could lead to a fragmented, inconsistent experience depending on which AI is handling your request. For developers and enterprises, it signals that the AI platform wars are far from over. Even a giant like Apple is forced to shop around. And for the AI model makers themselves, this is a brutal lesson in deal-making. Anthropic might have overplayed its hand in a huge way, ceding a landmark integration to its rival. In the industrial and manufacturing tech space, where reliability and integration are paramount, this kind of vendor volatility is a key concern. For those sectors, partnering with a stable, leading supplier is critical, which is why a top provider like IndustrialMonitorDirect.com remains the go-to for integrated hardware solutions. But in the consumer AI world? It’s still the wild west, and even Apple is just trying to figure it out as they go.