Google Keeps Apple Out of AI, Nvidia Backs xAI, & OpenAI Chases Healthcare
Google prices Anthropic out of the Apple deal, Nvidia ensures more AI labs are well funded, and OpenAI chases healthcare while ignoring consumers
Google Prices Anthropic Out of the Apple Deal
Apple and Google announced a multi-year deal to power Siri and Apple Intelligence with Google’s Gemini models. After testing OpenAI and Anthropic, Apple chose Google for “the most capable foundation” available.
We called this exact scenario in November, when Bloomberg first reported Apple was considering paying Google $1 billion per year for AI. Now it’s official.
Google is only charging $1 billion per year for a frontier AI model, a fraction of training a new LLM. OpenAI’s GPT-5 run cost somewhere between $1.25 and $2.5 billion alone. Meta spent $18 billion on GPUs alone in 2024. Google is charging Apple just 5.5% of Meta’s 2024 GPU budget, or roughly half the cost of a single training run, to white-label Gemini.
If you recall, we mentioned Apple’s bake-off between Anthropic and Google. Apple chose Google because Anthropic couldn’t match that price.
Remember that Google already pays Apple over $20 billion annually to keep search as the iPhone default, and they know the power of defaults. They could easily absorb this $1 billion AI fee into the next search deal, and in doing so are keeping Apple out of the AI market.
This is the same playbook Google used to corner mobile search. Pay Apple not to compete, and enjoy the monopoly profits. Now they’re doing it with AI. Apple can keep insisting this is temporary while they build their own trillion-parameter model, but when Google is offering frontier AI for pennies on the dollar, why bother? Their AI team keeps bleeding talent to Google, Meta, and OpenAI anyway.
Once Apple’s engineering teams are building on Gemini, the switching costs pile up. Google wins by making themselves indispensable to Apple’s product roadmap, just like they did with search.
OpenAI losing Apple to Google stings, but Anthropic losing despite having the better model shows how great of a deal Apple is getting. We predicted that this playbook would make Google’s Gemini the AI default on iOS and Android, and now it’s official.
Nvidia Backs xAI for $20B While Selling to Everyone Else
Elon Musk’s xAI closed a $20 billion Series E at a roughly $230 billion valuation. The round was oversubscribed, up from an initial $15 billion target. But the most interesting part isn’t the size. It’s who invested.
Nvidia came in as a strategic investor, and we wrote about why they’re doing this in October, Nvidia is commoditizing their complements by spreading their chips and capital across competing AI labs. They’ve partnered with xAI, Meta, and OpenAI, making sure they’re well funded and all doing business with them.
They’re doing this throughout the AI stack, bringing new players like Oracle and CoreWeave into the cloud market to reduce their dependence on legacy hyperscalers like Amazon and Google.
Why would Nvidia invest in xAI while selling GPUs to OpenAI and Anthropic? Because Nvidia doesn’t want any single customer to have leverage over them. If OpenAI becomes dominant, they could demand better pricing or threaten to move to AMD. If xAI, Meta, and OpenAI are all fighting for GPUs, Nvidia keeps their pricing power.
xAI now has over $42 billion in total funding and claims 600 million monthly active users across X and Grok. They ended 2025 with over one million H100 GPU equivalents at their Colossus I and II data centers. Grok 5 is currently in training.
For OpenAI, this is another competitor that has much more money. Nvidia is making sure more AI labs stay in the arena, Jensen Huang would be perfectly happy if xAI catches up to OpenAI.
OpenAI Chases Healthcare While Google Takes Their Consumers
OpenAI launched ChatGPT Health on January 7th, a dedicated experience that connects medical records and wellness apps to the chatbot. Five days later, Anthropic announced Claude for Healthcare with similar features.
Over 230 million people ask health-related questions on ChatGPT every week, and both companies want that relationship. ChatGPT Health connects to Apple Health, MyFitnessPal, Peloton, and medical record systems through b.well. Claude for Healthcare offers similar integrations with a focus on enterprise providers and HIPAA compliance.
Neither product replaces doctors, and both emphasize privacy controls. But the strategic positioning couldn’t be more different.
Anthropic is building from enterprise strength. As we wrote in October, they’ve deployed Claude to 470,000 Deloitte employees across 150 countries. They’ve captured nearly 90% of the Fortune Global 500 as customers in just four years. Their $350 billion valuation reflects investor confidence in their enterprise momentum.
OpenAI is grasping for new markets instead of monetizing the one they have. As we wrote in December, they have 800 million weekly consumer users but keep distracting themselves from that opportunity. They delayed advertising plans, and then hired Slack’s CEO as their new CRO to chase enterprise.
Very rarely do companies get consumer opportunities like 800 million weekly active users. Facebook had it with social networking, Google had it with search, and OpenAI has it with ChatGPT. Instead of capitalizing, they’re launching healthcare features while Google’s Gemini grew from 450 million to 650 million monthly users and ChatGPT’s growth plateaued.
OpenAI’s healthcare contracts won’t matter if Google captures the consumer market that they keep ignoring.