Skip to content
Longterm Wiki
Back

China’s AI disrupter DeepSeek bets on low-key team of ‘young geniuses’ to beat US giants | South China Morning Post

web
scmp.com·scmp.com/tech/big-tech/article/3294357/chinas-ai-disrupte...

This article covers DeepSeek's emergence as a competitive AI lab using a small team of young researchers, relevant to AI safety discussions about capability diffusion, compute efficiency, and the geopolitical dynamics of AI development.

Metadata

Importance: 52/100news articlenews

Summary

DeepSeek, a Chinese AI startup spun off from hedge fund High-Flyer Quant, released its V3 large language model in December 2024, matching or exceeding US competitors despite limited compute resources. The company relies on a small team of roughly 150 researchers, preferring fresh graduates over experienced hires. Its breakthrough raises questions about whether chip restrictions can effectively limit China's AI progress.

Key Points

  • •DeepSeek V3 was developed by ~150 researchers and matched or exceeded performance of models from Meta and OpenAI despite fewer resources.
  • •The company was spun off in 2023 from High-Flyer Quant, a Chinese hedge fund, by founder Liang Wenfeng.
  • •DeepSeek deliberately hires fresh graduates or early-career researchers, prioritizing raw ability over experience.
  • •The model's efficiency challenges assumptions that US chip export controls can effectively constrain Chinese AI capabilities.
  • •Key architectural innovations (MLA architecture) were credited to specific young researchers, highlighting the team's technical depth.

2 FactBase facts citing this source

EntityPropertyValueAs Of
DeepSeekHeadcount100Jul 2023
DeepSeekHeadcount150Jun 2024

Cached Content Preview

HTTP 200Fetched Apr 7, 20263 KB
China’s AI disrupter DeepSeek bets on low-key team of ‘young geniuses’ to beat US giants | South China Morning Post Advertisement Artificial intelligence Tech Big Tech China’s AI disrupter DeepSeek bets on low-key team of ‘young geniuses’ to beat US giants 

 DeepSeek prefers to hire new graduates, or those early in their AI career, in line with the company’s preference for ability over experience

 3 -MIN READ 3 -MIN 1 Ben Jiang in Beijing Published: 9:22am, 12 Jan 2025 Updated: 10:32am, 28 Jan 2025 DeepSeek, the Chinese artificial intelligence (AI) start-up that took the tech world by surprise with its powerful AI model developed on a shoestring, is betting on its secret weapon of “young geniuses” to take on deep-pocketed US giants, according to insiders and Chinese media reports.

 On December 26, the Hangzhou-based firm released its DeepSeek V3 large language model (LLM), which was trained using fewer resources but still matched or even exceeded in certain areas the performance of AI models from its larger US competitors such as Facebook parent Meta Platforms and ChatGPT creator OpenAI . The breakthrough is considered significant as it could offer a path for China to exceed the US in AI capabilities despite its restricted access to advanced chips and funding resources. DeepSeek did not immediately respond to a request for comment on Friday.

 The DeepSeek logo is displayed on a smartphone. Photo: Shutterstock Images Behind its breakthrough is the firm’s low-key founder and a nascent research team, according to an examination of authors credited on its V3 model technical report and career websites, interviews with former employees, as well as local media reports. The V3 technical report is attributed to a team of 150 Chinese researchers and engineers, in addition to a 31-strong team of data automation researchers.

 Advertisement The start-up was spun off in 2023 by hedge-fund manager High Flyer-Quant . The entrepreneur behind DeepSeek is High-Flyer Quant founder Liang Wenfeng, who studied AI at Zhejiang University. Liang’s name is also on the technical report. In an interview with Chinese online media outlet 36Kr in May 2023, Liang said most developers at DeepSeek were either fresh graduates, or those early in their AI career, in line with the company’s preference for ability over experience in recruiting new employees. “Our core technical roles are filled with mostly fresh graduates or those with one or two years of working experience,” Liang said.

 Advertisement Among DeepSeek’s breadth of talent, Gao Huazuo and Zeng Wangding are singled out by the firm as having made “key innovations in the research of the MLA architecture”.

 Advertisement Select Voice Select Speed 0.8x 0.9x 1.0x 1.1x 1.2x 1.5x 1.75x 00:00 00:00 1.00 x
Resource ID: kb-0770a1930b611267