Meta Develops Specialized AI Chip, Reports Suggest

Meta is testing its own chip for training AI systems, according to Reuters, citing sources.

The agency reports that the new processor is a specialized accelerator designed to tackle specific artificial intelligence tasks. This approach renders the chip more energy-efficient compared to integrated graphics processors traditionally used for AI workloads.

The company is collaborating with Taiwan’s TSMC. The initial development phase, which includes the creation and dispatch of prototypes to the chip manufacturing plant for testing, has been completed.

The development of proprietary processors is part of Meta’s strategy to reduce infrastructure costs. The company is betting on artificial intelligence to drive growth.

The corporation forecasts expenditures for 2025 to be between $114–119 billion, with $65 billion allocated to the artificial intelligence sector.

Meta aims to begin deploying its own chips for AI tasks in 2026, Reuters writes.

In May 2023, the company introduced two specialized processors for artificial intelligence and video processing tasks.

Previously, media reported on OpenAI’s work on its own AI processor in partnership with Broadcom and TSMC.

Chinese company ByteDance, in collaboration with Broadcom, is developing a similar product.