Apple is joining forces with Broadcom to create its first AI-focused server chip, designed to meet the growing demands of the company’s AI-driven features, according to a report from The Information.
The new chip, internally known as “Baltra,” is expected to begin mass production by 2026. This move follows a similar path taken by other tech giants such as Google, Amazon, and Meta, all of which have developed custom chips tailored to their AI services.
This initiative is part of Apple’s broader push into AI, which includes tools for generating text, creating images, and summarising content. While Apple’s existing chips are capable of powering some AI tasks, they are not as optimised as chips that are specifically designed for AI processing.
Broadcom, known for its work with Google on AI chips, will contribute networking technology to the project. However, Apple will maintain control over the overall design and production of the chip.
Apple has a strong track record of in-house chip development, having replaced Intel processors with its own M-series chips in Mac computers. However, despite these successes, NVIDIA continues to dominate the AI chip market, though Google has been able to reduce its reliance on the company by collaborating with Broadcom.
In another sign of the industry’s shift toward custom chips, OpenAI—the company behind ChatGPT—has partnered with Broadcom and TSMC to produce its first AI chip, also expected in 2026.
Apple’s move to develop custom chips is part of a larger strategy to enhance the security and efficiency of AI data processing. As the company continues to roll out more AI-powered features, the Baltra chip will likely play a crucial role in scaling services to millions, if not billions, of devices.