Apple is widely recognized for designing world-class silicon for iPhones and Mac. However, running heavy-duty artificial intelligence in the cloud is proving to be an entirely different beast. According to a report, the iPhone maker is actively exploring the acquisition of semiconductor startups. The company has reportedly held discussions with investment bankers and approached several chip designers to gauge their interest in a buyout, aiming to aggressively bolster its AI server infrastructure.
This sudden hunt for chip acquisitions highlights a rare vulnerability in Apple’s hardware pipeline. The Information claims the company is currently facing performance bottlenecks with its custom in-house AI servers, which rely on M2 Ultra processors. When Apple recently attempted to run Google’s massive Gemini models on these internal servers to power its revamped, AI-driven Siri, the Mac-based chips simply could not handle the processing load. As a result, Apple had to outsource the heavy lifting. They are running those demanding tasks on Nvidia GPUs hosted within Google’s cloud infrastructure (via Engadget).
The delayed roadmaps
Apple originally planned to debut a custom server processor, code-named “Baltra,” to solve these performance issues. However, sources familiar with the matter say the project has slipped past its initial release window. Furthermore, analysts suggest that a true successor, potentially based on an M7 Ultra chip, is not expected to be ready until 2029.
To bridge this massive gap, Apple is temporarily upgrading its infrastructure with newer M5 Ultra chips. The company also recently locked in a massive $30 billion multi-year supply deal with Broadcom to secure its domestic chip sourcing. But to truly gain independence in the cloud AI space, Apple needs specialized server silicon expertise, which is driving its search for an immediate acquisition.
Apple notoriously avoids multi-billion-dollar buyouts—with its $3 billion purchase of Beats in 2014 remaining its largest. However, the company has shown a growing willingness to spend big on AI. It acquired Israeli AI audio startup Q.ai for nearly $2 billion earlier this year, and recently held talks with PrismML, a firm specializing in shrinking large AI models. With over $45 billion in cash reserves and a recent shift in financial policy that gives the company more purchasing flexibility, a major semiconductor acquisition is highly feasible.
The post Apple is Quietly Shopping for AI Chip Startups as M2 Ultra Isn’t Cutting It appeared first on Android Headlines.
