ByteDance, the company behind TikTok, is starting to develop its own artificial intelligence chips. ByteDance currently relies heavily on NVIDIA for AI, but plans to begin mass production of its own AI chips by 2026.

The company will produce The new chips will be used for two purposesWhile AI training involves teaching a machine learning model large amounts of data, AI inference will involve a model trained to make real-time predictions, such as TikTok’s personalized content recommendations.

Why is ByteDance going to produce its own artificial intelligence chips?

ByteDance to Buy Around 200,000 NVIDIA H20 GPUs This Year More than 2 billion dollars are spent That has happened, but most of them have not yet been delivered. Due to the high costs and limited supply, the company wants to reduce its dependence on NVIDIA by creating its own artificial intelligence chips. To achieve this, ByteDance is working with Broadcom, an experienced AI chip designer, and will use technologies from TSMC, a leading chipmaker.

While ByteDance’s GPUs aren’t as powerful as NVIDIA’s top-end models, the company hopes the move will help it save money and give it more control over its technology. However, the company is primarily interested in developing a new software platform that could replace NVIDIA’s CUDA, which is needed to run artificial intelligence systems. also faces major challenges.

If ByteDance is successful, it could fundamentally change China’s AI hardware market and reduce the country’s dependence on foreign vendors like NVIDIA. In the long run, however, it will take time to compete with NVIDIA, the undisputed leader in artificial intelligence hardware.

So what do you think about this issue? You can share your opinion with us in the comments section below.