US$1 Trillion Wipeout – How China’s DeepSeek AI Creates Havoc And Spooks U.S. Financial Markets

US$1 Trillion Wipeout – How China’s DeepSeek AI Creates Havoc And Spooks U.S. Financial Markets



January 29th, 2025 by financetwitter



For decades, China has been accused by the United States and several of its allies of economic espionage and intellectual property (IP) theft. It was claimed that Chinese IP theft has cost the U.S. about US$225 billion to US$600 billion a year. Besides stealing IP, China was also accused of demanding technology transfers in exchange for market access.

However, if it’s true that China only knows how to steal and does not know how to innovate and invent, exactly who invented paper-making, printing, gun-powder and the compass in the first place? Other ancient Chinese inventions included alcohol, mechanical clock, tea production, umbrella, acupuncture, iron smelting, porcelain, silk, bronze, kite, seed drill, and even earthquake detector.

And if indeed the U.S. is leading the technology, why were they so afraid of Huawei’s 5G technology and Chinese EVs (electric vehicles) that the Americans are instigating – even threatening – allies to boycott, block and ban the Chinese 5G and EVs from entering their markets? Hilariously, Biden and Trump were trying to ban WeChat and TikTok too,despite claims of Chinese IP theft.


Yes, naughty China had stolen so much Western technology that the Chinese have become not only the leader in 5G and EVs, but also in short-video and hypersonic missile. The U.S., supposedly the leader in virtually every pillar, experienced another “Sputnik moment” when its artificial intelligence (AI) leadership was challenged by Chinese DeepSeek on Monday (January 27).

The sudden emergence of DeepSeek, a competitor to OpenAI and its ChatGPT tool, saw US$1 trillion wiped out from Nasdaq Composite in one day. Even Nvidia, a leading maker of the computer chips that power AI models, suffered the biggest fall in U.S. stock market history when its shares fell 17% – wiping nearly US$600 billion, while Google’s parent company lost US$100 billion and Microsoft US$7 billion.

Powered by the open source DeepSeek-V3 model, DeepSeek, a Chinese AI chatbot was made at a fraction of the cost (US$6 million) – significantly less than the billions spent by rivals. Even though it was launched last week, it has already become the most downloaded free app in the U.S. and UK, raising questions about the future of America’s AI dominance. But it was just the beginning.


The arrival of DeepSeek was shocking as the U.S. thought it has successfully crippled China’s technology by restricting the sale of the advanced chip technology that powers AI to China. But without supplies of imported advanced chips, Chinese AI developers were forced to share their work with each other and experimented with new approaches to the technology. This has resulted in AI models that require far less computing power than before.

Nvidia’s most advanced chips – H100s – have been banned from export to China since September 2022 thanks to US sanctions. Nvidia then developed the less powerful H800 chips for the Chinese market, but they were also banned from export to China last October. DeepSeek’s success at building an advanced AI model without access to the most cutting-edge US technology means the sanctions have backfired.

Crucially, it also means that they cost less than previously thought possible, which could create chaos in the industry. The company, founded by entrepreneur Liang Wenfeng who runs hedge fund “High-Flyer Capital and uses AI to identify patterns in stock prices, said that DeepSeek-R1 was of “performance on par with” one of OpenAI’s latest models when used for tasks such as maths, coding and natural language reasoning.


Initially, 40-year-old Liang bought Nvidia chips in 2021 to develop AI models as a hobby, bankrolled by his hedge fund. In 2023, he founded DeepSeek in the eastern Chinese city of Hangzhou. The company is purely focused on research rather than commercial products – DeepSeek and its codes can be downloaded for free, while DeepSeek’s models are also cheaper to operate than OpenAI’s.

Caught off-guard by their rival’s technological achievement, the U.S. was both unimpressed and upset. DeepSeek now raises doubts over the necessity for hefty investment in AI infrastructure such as chips and the so-called market-leading role of US tech companies in AI. Analysts at US investment bank Goldman Sachs already raised the alarm over AI spending last year in a note – “Gen AI: too much spending, too little benefit?”

US President Donald Trump, who used to belittle Chinese technology, said DeepSeek should be a “wake-up call for our industries that we need to be laser-focused on competing to win”. Meanwhile, Silicon Valley venture capitalist and Trump adviser Marc Andreessen described DeepSeek-R1 as “AI’s Sputnik moment”, a reference to the satellite launched by the Soviet Union in 1957 that shocked the U.S.


However, OpenAI chief executive Sam Altman praised DeepSeek, saying that it was “invigorating to have a new competitor” and called it “an impressive model, particularly around what they’re able to deliver for the price”. The European as well as the Japanese tech firms linked to the AI sector also tanked, equally spooked by the Chinese AI chatbot.

While some experts have played down DeepSeek, the fact that it was targeted in a cyber-attack on Monday, forcing it to temporarily limit registrations, suggests that certain people in Washington were extremely terrified of the Chinese AI technology. Essentially, the U.S. can’t control China as the belief that whoever wields the world’s most advanced semiconductors controls the AI race is no longer true.

DeepSeek appears to have accomplished its feat using open-sourced generative AI to build and teach its LLM (Large Language Model). It used a process called “fine-tuning” on a large dataset of text interactions gathered from ChatGPT’s existing open-source model. And it did so far more cheaply, and far more quickly than OpenAI and others who have built their own machine-learning models.


Unhappy with DeepSeek’s accomplishment, OpenAI has accused the Chinese artificial intelligence start-up of using the U.S. company’s proprietary models to train its own open-source competitor, therefore breaching OpenAI’s IP. However, OpenAI is battling allegations of its own copyright infringement from newspapers and content creators, who accuse the company of training its models on their articles and books without permission..

More importantly, it diminishes the geopolitical leverage that Washington and its allies (such as Japan, South Korea, Taiwan and Europe) wield through export controls, sanctions and other restrictions on tech transfer. This means even with second-tier semiconductor, China can achieve first-tier AI status and enjoy benefits in military prowess and economic competitiveness, dealing a blow to the U.S.