Foreign Media: Threat of Anthropic Mythos Prompts DeepSeek to Secure First Massive Funding

According to Dongcha Beating monitoring, The Information reported citing sources familiar with the matter that DeepSeek has decided to abandon the founder's full investment model and initiate a first round of funding exceeding 500 billion RMB (approximately $7.4 billion), with the direct trigger being Anthropic's security model Mythos showcased in April.

Mythos has the capability to discover and exploit software vulnerabilities. DeepSeek's R&D team assessed that in the face of a model requiring massive computing power and data stacking, potentially even possessing a nation-state-level destructive capability, relying solely on Liang Wenfeng's personal wealth would not be able to compete in the computing power arms race, necessitating the introduction of external funding to expand the computing power pool.

To address US export controls last year, DeepSeek attempted to migrate its training and deployment system from Nvidia's CUDA ecosystem to Huawei chips, rewriting the underlying software, which resulted in a 15-month delay in the release of the new generation model.

DeepSeek has maintained an extremely flat operation in the past, with no human resources or public relations department established, and everyone reporting directly to Liang Wenfeng. With the completion of a huge financing round, the team is now beginning to recruit functional positions such as legal, financial, and human resources.