Why DeepSeek delayed launch of its V4 AI model
What's the story
DeepSeek, a Chinese artificial intelligence (AI) start-up, delayed the launch of its much-anticipated V4 model. The company had focused on optimizing the software stack for Huawei's Ascend chips. The decision comes as part of a larger effort to strengthen China's domestic chip industry and also reduce dependence on foreign hardware.
Strategic alignment
China's push for technological self-reliance
DeepSeek's decision to delay the V4 model launch aligns with China's broader goal of technological self-reliance. The country is investing in domestic solutions, even if they don't match global top-tier performance yet, to build a more resilient and independent AI ecosystem. This strategy comes amid increasing restrictions on access to foreign chips, further emphasizing the need for homegrown alternatives.
Market competition
Preview versions released
Despite the delay, DeepSeek has released preview versions of its new flagship AI model. The company claims it is the "most powerful open-source platform" yet, taking on rivals like OpenAI and Anthropic. Industry analysts say that the growing adoption of domestic semiconductors and increased capital support have permitted early deployments of large language models trained and run entirely on local compute.