Apple is reportedly exploring the acquisition of semiconductor startups to enhance its AI server infrastructure. Recent discussions with bankers and chip companies indicate the tech giant’s intent to strengthen its capabilities in this area.
Currently, Apple’s AI servers operate on internally designed M2 Ultra chips. However, the development of its next-generation server chip, code-named Baltra, has been delayed beyond the planned 2026 release. To manage more demanding AI tasks, Apple relies on the Gemini-based model powering the new Siri, which utilizes Nvidia GPUs within Google Cloud.
Historically, Apple has made significant acquisitions to bolster its technological portfolio. In 2014, the company acquired Beats for $3 billion, and in 2019, it purchased the majority of Intel’s smartphone modem business for $1 billion. More recently, Apple acquired Israeli AI startup Q.ai for nearly $2 billion, reflecting its ongoing commitment to advancing AI capabilities.
Additionally, Apple has engaged in discussions with PrismML, a company specializing in compressing large AI models to run efficiently on devices like the iPhone. While no formal deal has been announced, this move aligns with Apple’s strategy to enhance on-device AI processing.
These developments coincide with a shift in Apple’s financial strategy. The company’s CFO, Kevan Parekh, indicated that achieving a net cash-neutral position is no longer a formal goal, suggesting increased flexibility for larger acquisitions.
Apple’s pursuit of AI chip startups underscores its dedication to advancing AI infrastructure. By integrating specialized hardware, Apple aims to enhance the performance and efficiency of its AI services, positioning itself competitively in the rapidly evolving AI landscape.
