Apple’s highly anticipated, in-house AI server processor, code-named “Baltra,” has been delayed. Originally targeted to ship this year to power the company’s Private Cloud Compute infrastructure, the custom silicon project has slipped.
The reported delay comes as Apple faces mounting pressure to strengthen the infrastructure behind its AI ambitions. According to a report by The Information, Apple is grappling with severe performance limitations on its current AI servers, which rely on Mac-based M2 Ultra chips.
While highly efficient for personal computers, these chips have proven insufficient when tasked with running complex, large-scale generative AI models.
Handing the reins to rivals
The reported delay also sheds light on why Apple has been racing to develop dedicated AI server hardware. According to The Information, the company’s existing infrastructure has struggled to keep pace with the demands of modern generative AI workloads.
The performance gap became glaringly obvious when Apple engineers attempted to run Google’s Gemini models locally on Apple’s own servers to power a revamped Siri. The Information reported that the hardware was unable to deliver the performance Apple needed for the project.
Apple reportedly turned to Nvidia GPUs hosted on Google’s cloud infrastructure for the compute-intensive portions of its upgraded Siri effort. For a company that fiercely guards its independence and markets user privacy through local processing, relying on rival clouds and Nvidia hardware is a significant strategic compromise.
Breaking the M&A playbook
To close this technological gap, Apple is reportedly breaking its decades-long pattern of conservative mergers and acquisitions. The company approached semiconductor startups and consulted with investment bankers about potential buyouts to accelerate its server silicon timeline.
The reported strategy aligns with broader changes in Apple’s approach to capital deployment. CFO Kevan Parekh recently signaled that Apple is moving away from its long-held “net cash neutral” policy, freeing up its massive $45.6 billion cash pile for larger acquisitions, BigGo Finance reported.
The shift is already underway. Earlier this year, Apple completed its reported acquisition of Israeli AI startup Q.ai for nearly $2 billion, expanding its AI capabilities. Taken together, the reported moves suggest Apple is increasingly willing to supplement its in-house chip development with acquisitions as it works to close its AI infrastructure gap.
Must-read Apple coverage
- Nearly 7 in 10 iPhone Owners Plan iPhone 17 Upgrade
- Apple Prepares AI-Powered Siri Upgrade With Google Search Integration
- Apple’s Xcode 26 Beta Now Supports GPT-5 and Claude
- Apple’s Vision Pro Adoption Stalled As Content Released ‘Drip by Drip’
The architectural chasm of the cloud
Apple’s struggle highlights a profound engineering friction. For nearly two decades, Apple’s chip design team has mastered the art of efficiency-per-watt, optimizing low-power silicon to prevent iPhones and MacBooks from overheating in users’ hands.
However, AI training and high-concurrency server inference require the exact opposite: brute-force thermal design, massive memory bandwidth, and complex interconnect fabrics capable of chaining thousands of processors together. Apple is learning that stacking consumer-grade Mac chips in a server closet cannot replicate a purpose-built AI data center.
Risks and what this means for consumers
For consumers, these delays mean the highly anticipated, fully local Siri overhaul may remain hybrid and dependent on external networks longer than expected. For investors, the tradeoff is that Apple’s capital expenditures will rise sharply.
Buying chip startups is highly expensive in a seller’s market, and integrating foreign semiconductor architectures into Apple’s proprietary ecosystem introduces immense execution risk. If Apple cannot deliver its custom server silicon soon, its reliance on Nvidia and Google will only deepen, threatening both its profit margins and its core privacy narrative.
More News: Apple Intelligence has cleared a major regulatory hurdle in China, paving the way for an AI rollout powered by Alibaba and Baidu.