Apple is intentionally avoiding the massive AI capital expenditure race, betting that foundation models and the underlying compute will become commoditized. This strategy allows them to wait for the market to mature and prices to drop before integrating the technology, rather than spending billions to build their own models from behind.
Apple's ultimate advantage in the age of AI may be its hardware ecosystem, particularly the iPhone. As the central computing hub for billions of users, the iPhone is perfectly positioned to be the primary device for running on-device models and AI applications, ensuring Apple's relevance regardless of who builds the best foundational AI.
Contrary to the popular narrative that open-source AI will quickly commoditize the market, there is evidence that the frontier is accelerating faster than the open-source community can keep up. This potential divergence challenges the 'good enough' argument and suggests that proprietary models may maintain a significant, defensible lead for longer than expected.
While Microsoft now allows OpenAI to partner with rivals like AWS, it secretly retained a powerful advantage. A new clause mandates that all OpenAI products must launch first on Azure, unless Microsoft is unable or unwilling to support them. This gives Microsoft an exclusive 'theatrical release window' for the most advanced AI models.
The greatest threat to Apple's AI strategy is a future where a single, all-powerful AI assistant becomes the primary user interface. In this scenario, the underlying AI model is the only thing that matters, potentially rendering Apple's hardware a commoditized shell and forcing them to license the core technology from a competitor like OpenAI or Google.
The race to build frontier AI models is not just about capital. Despite enormous investment, companies like Amazon (with its Nova model), Meta, and xAI have failed to catch up to the leaders. This suggests that talent, timing, and research culture are critical variables that money alone cannot solve, potentially validating Apple's decision to stay on the sidelines.
Incoming CEO John Ternus is already preparing to shift away from Tim Cook's signature policy of returning cash to shareholders via buybacks. By signaling this change early, Ternus is creating financial optionality to make major acquisitions, ramp up R&D, or significantly increase AI infrastructure spending without shocking Wall Street.
OpenAI's ambitious Stargate initiative has quietly pivoted from a strategy of building and owning its own massive AI infrastructure to one of securing capacity from partners. This move de-risks OpenAI's balance sheet but transfers the immense financial and operational risk onto its infrastructure partners, whose business models now depend heavily on OpenAI's continued demand.