Apple Surges Past Nvidia as AI Capex Outlook Darkens
Apple's brief reclaiming of the global valuation crown from Nvidia signals a market shift away from massive data center spending toward cheaper, on-device AI.
On July 17, Apple briefly unseated Nvidia to become the world's most valuable company, a position it had not held since April of last year. The shift reflects a broader market reassessment of artificial intelligence, moving away from companies dependent on massive data center expenditures.
The rally in AI chipmakers is losing momentum as investors question whether hyperscalers can generate adequate returns to justify the nearly $1 trillion AI spending cycle. UBS estimates capital expenditures will surge 76% this year to $673 billion. However, growth is projected to decelerate sharply to 25% next year and a mere 6% by 2028. Having initially funded the infrastructure buildout with cash reserves, hyperscalers are now turning to external financing. This shift raises significant concerns about capital market pressures and the sustainability of spending growth.
Downward pressure on AI pricing is intensifying with the arrival of low-cost competition from China. Beijing-based startup Moonshot AI recently launched Kimi K3, a 2.8-trillion-parameter model that claims to close the gap with US offerings, even surpassing systems from OpenAI and Anthropic on certain benchmarks.
Morgan Stanley analyst Gary Yu noted the release signals a broader industry shift. "K3 has received positive feedback globally, signaling an all-round catch-up of Chinese LLMs with US leaders in model size, performance, and pricing. K3, the largest open-weight Chinese LLM so far, may suggest the scaling law still holds... We do not view K3 as an overnight miracle but rather as the result of cumulative progress across China's AI model industry."
As capable models become substantially cheaper, enterprises are likely to deploy low-cost systems for routine tasks and reserve expensive proprietary infrastructure for complex workloads. This bifurcation threatens the pricing power of model developers and cloud providers, making it harder to generate attractive returns on costly infrastructure investments.
This changing landscape highlights the relative strength of Apple's hardware-centric approach, which relies on on-device processing rather than vast server farms. "Apple was seen as a laggard in the AI race because it wasn't spending to develop models, but now sentiment has changed," said Toni Meadows, head of investment at BRI Wealth Management. For investors, the divergence underscores a new consensus: the next phase of AI will be defined by cost efficiency rather than sheer computational scale.
