Apple is collaborating with Broadcom to develop its first server chip designed specifically for artificial intelligence tasks, according to The Information. The chip, internally code-named Baltra, is anticipated to enter mass production by 2026, signaling a bold step in Apple's pursuit of advanced AI capabilities and its strategy to reduce reliance on external chip suppliers like Nvidia.
The Baltra chip will reportedly utilize Taiwan Semiconductor Manufacturing Co.'s (TSMC) cutting-edge N3P manufacturing process, reflecting Apple's commitment to deploying the latest technologies in its hardware development. TSMC has long been a key partner for Apple, enabling the production of its advanced processors, including the M-series chips powering its Mac devices.
While Apple and Broadcom have not publicly commented on the partnership, Broadcom's role in this project builds on its existing relationship with Apple. The companies previously entered a multi-billion-dollar agreement for the development of 5G radio frequency components, showcasing their close collaboration in hardware innovation.
The move underscores Apple's broader push into the AI market, aligning with other tech giants like Google, which also works with Broadcom for its AI chips. Companies in the industry are increasingly focused on creating in-house solutions to manage the growing computational demands of AI technologies, such as generative AI models and large-scale machine learning applications.
Apple's venture into AI-specific server chips could bolster its Private Cloud Compute system, which currently powers AI features like Apple Intelligence. These chips, optimized for server-based workloads, differ from the neural engines found in Apple's M-series chips designed for consumer devices. Such advancements are expected to elevate Apple's AI infrastructure, potentially enabling more sophisticated applications.
One area likely to benefit from this development is Siri. Rumors suggest Apple is working on a more conversational and intelligent version of the virtual assistant, similar to ChatGPT and other generative AI platforms. According to a Bloomberg report, this revamped Siri could launch as early as 2026, coinciding with the release timeline for the Baltra chip.
The partnership with Broadcom is also significant as Apple aims to decrease dependency on Nvidia, whose GPUs dominate the AI processing market but are often costly and difficult to source due to high demand. Developing proprietary chips could give Apple greater control over its AI ecosystem while addressing supply chain challenges.
Broadcom, already a key player in the AI chip market, has seen its stock value rise significantly amid the generative AI boom. In 2024 alone, its shares have gained 54%, building on last year's near-doubling in value. This momentum highlights the growing importance of AI hardware in driving technological innovation and market growth.
The global custom chip market is expected to grow significantly in the coming years, with projections of reaching $45 billion by 2028. Apple's collaboration with Broadcom positions it as a serious contender in this competitive landscape, joining the ranks of other companies investing heavily in AI-focused hardware.
