Last Monday, a small, low-profile Chinese AI lab called DeepSeek unveiled a set of artificial intelligence models so efficient they ma ke Silicon Vall ey’s best look outdated. These models are fifty times more efficient than even top U.S. offerings. The announcement has thrown American tech giants like OpenAI, Google, and Meta into a full-blown crisis, as their closed-door strategies suddenly stop cutting it. Now everyone on X is mocking them. For those confused why the US state and Silicon Valley are having a meltdown on twitter today. China released multiple AI models that are 50x more efficient than the best American AI models and made them open source, ruining the AI market — ashok kumar 🇵🇸 (@broseph_stalin) January 25, 2025 The man behind this shockwave is Liang Wenfeng, a hedge fund manager who turned a side project into one of the most disruptive AI advancements in history. DeepSeek’s R1 model is a self-learning system that can improve on its own without human oversight. And here’s what’s making Silicon Valley sweat: Liang is leveling the playing field for anyone—whether in China or anywhere else—looking to jump into AI development. Liang Wenfeng’s long game In 2021, while everyone else in China’s AI scene was focused on big tech, Liang quietly bought thousands of Nvidia graphic processing units (GPUs) and started experimenting with AI. People in the industry thought he was just another billionaire chasing a hobby. According to a business partner who reportedly spoke to the Financial Times, “When he told us he wanted to build a 10,000-chip cluster, we thought he was crazy. He didn’t even explain why, just said, ‘This will change everything.’” Fast forward to 2023, and Liang launched DeepSeek, hiring top AI engineers straight from his hedge fund. Using profits from the hedge fund High-Flyer, he built a team that not only understood AI but also mastered the infrastructure behind it. By 2024, DeepSeek had developed R1, a language model that industry insiders describe as a direct challenge to every major AI player in the U.S. Unlike its American competitors, DeepSeek wasn’t interested in commercialization. It focused entirely on research, with Liang using his own money to fund the operation. “DeepSeek’s offices feel like a university lab,” said the Financial Times. Based in Beijing and Hangzhou, the lab employs some of the best AI minds in China, offering salaries that rival TikTok-owner ByteDance. Liang’s singular goal, according to the report, was to prove that China could innovate on the same level as the U.S., and now he has done it. The timing couldn’t have been more deliberate either. Liang’s release of the R1 model coincided with his appearance at a high-profile meeting in Beijing led by Li Qiang, China’s second most powerful politician. Liang was the only AI leader invited. Li told the entrepreneurs present to focus on breaking through key technologies — a clear nod to China’s ambition to outpace the U.S. in AI development. Something tells us US president Donald Trump is not too happy about that. Silicon Valley’s scramble to catch up American tech giants, caught off guard, are now scrambling to respond. OpenAI, the company behind ChatGPT, announced a $100 billion joint venture with Japan’s SoftBank, dubbed Stargate, aimed at building new AI infrastructure in the U.S. Meanwhile, Elon Musk’s xAI is expanding its Colossus supercomputer, planning to deploy over one million GPUs to train its Grok AI models. Google, Meta, and Anthropic are also throwing billions into upgrading their computing clusters with Nvidia’s next-generation Blackwell chips. But the U.S. companies have one major disadvantage: secrecy. For years, Silicon Valley has operated on a closed-door model, keeping AI breakthroughs locked behind proprietary systems. DeepSeek’s decision to make R1 open source has flipped that narrative. This has triggered a wave of panic in the U.S., where companies now face pressure to decide whether to follow DeepSeek’s lead. But there are also concerns about whether DeepSeek can maintain its momentum. Despite its success, the company’s resources are admittedly limited compared to U.S. giants. “They’ve built one of the largest computing clusters in China,” said a source familiar with the company. “But compared to what OpenAI and Google are building, it’s not enough. They’ll need to scale up if they want to keep up.” Cryptopolitan Academy: How to Write a Web3 Resume That Lands Interviews - FREE Cheat Sheet
