Chinese artificial intelligence company DeepSeek is reportedly developing its own artificial intelligence chip, marking a significant step towards reducing its dependence on external suppliers such as Nvidia and Huawei while strengthening its long-term semiconductor strategy.

According to reports citing people familiar with the matter, the company has been working on the project for around a year and is focusing on a processor designed specifically for AI inference—the stage at which trained models generate responses to users—rather than for training large language models.

If successful, the initiative would represent a major strategic shift for DeepSeek, which has gained international recognition for developing high-performance AI models despite operating under increasingly restrictive United States export controls on advanced semiconductor technology. The move also reflects a broader industry trend in which leading AI developers are investing in proprietary silicon to improve performance, lower operating costs and reduce reliance on third-party hardware suppliers.

Sources indicate that DeepSeek has been expanding its semiconductor design capabilities by recruiting chip engineers and engaging with potential foundry, memory and chip design partners. The company is believed to be exploring manufacturing options with Semiconductor Manufacturing International Corporation (SMIC), although no official details regarding fabrication have been confirmed.

The reported chip is expected to target the rapidly growing inference market, where demand is accelerating as artificial intelligence applications move from model training to commercial deployment. Inference processors are generally designed to deliver lower power consumption and greater efficiency than general-purpose graphics processing units, making them increasingly attractive for large-scale AI services.

DeepSeek currently relies on a combination of Nvidia and Huawei hardware to train and deploy its AI models. However, tightening United States export restrictions on advanced AI accelerators have encouraged Chinese technology companies to accelerate investment in domestic semiconductor capabilities and custom chip development.

The company's reported move also places it alongside other major artificial intelligence developers and global cloud service providers that are increasingly designing proprietary silicon to optimise artificial intelligence infrastructure and reduce long-term hardware costs.

Although the project is still in its early stages and commercial production has not been confirmed, DeepSeek's reported investment in chip design highlights China's continued drive to strengthen domestic semiconductor capabilities amid ongoing geopolitical tensions and export restrictions. If successful, the initiative could enhance the company's control over its artificial intelligence infrastructure while supporting the broader development of China's semiconductor ecosystem.