As the festivities of the Lunar New Year brought families together, a revolutionary AI platform developed in China captured the attention of both domestic and international audiencesThe emergence of DeepSeek, a pioneering product born out of a Chinese startup, has stirred excitement in the AI industry, showcasing exceptional performance combined with costs that challenge conventional norms.
On January 27, when the American stock market opened, it witnessed a significant downturn, particularly in the tech sectorNotably, Nvidia, a titan in the chip industry, saw a staggering 17% drop in its share priceOther companies like Broadcom and AMD reported similar declines of 17% and 6% respectively, while Microsoft’s stock fell by 2%. The widespread effects even touched utility companies linked to artificial intelligence, as evidenced by the 21% decline in stock prices for the American Electric Power Company and a 29% fall for Vistra.
The launch of DeepSeek did not just send shockwaves through the American markets but also generated ripples in the domestic scene
Advertisements
Its parent company, Daily Interactive, hit the daily limit on the last trading day before the holiday, achieving a remarkable 20% surge within a weekThis phenomenon was largely attributed to the positive sentiment surrounding DeepSeek and its innovative solutions.
Since its initial launch, the DeepSeek platform has sparked a steady increase in the AIGC (Artificial Intelligence Generated Content) index within the Hong Kong stock markets, with data indicating a cumulative rise of 27.23% over eight trading days beginning January 21.
The buzz surrounding DeepSeek did not dwindle during the Lunar New Year; in fact, the platform made headlines by offering salaries of up to a million dollars for recruitment, further enhancing its visibilityDespite facing malicious attacks aimed at tarnishing its reputation, the platform garnered significant user support both domestically and abroad.
According to reports from international investment banks, DeepSeek's valuation has skyrocketed to an astonishing 10 billion USD
Advertisements
This figure reflects not only the platform’s intrinsic value but also the market reaction indicating a broader shift within the AI industry.
DeepSeek, formally known as Hangzhou DeepSeek Artificial Intelligence Basic Technology Research Co., Ltd., was founded on July 17, 2023. The company is dedicated to developing cutting-edge large language models (LLMs) and associated technologiesWith a small yet dedicated founding team, DeepSeek has managed to create an open-source model that rivals those of established players like OpenAI.
The founder, Liang Wenfeng, is a remarkable figure born in the 1980sHe achieved academic accolades early on, becoming one of the top scorers in the national college entrance examination at the age of 17 and enrolled in Zhejiang University, where he later pursued postgraduate studiesIn 2013, he co-founded a quantitative investment company with the goal of leveraging mathematics and AI for financial gains, establishing a strong base for his later venture into the AI industry.
By 2021, his previous venture had amassed over 100 billion RMB in asset management
Advertisements
He made a bold move into the realm of general artificial intelligence in 2023 by launching DeepSeek, focusing on developing truly human-level AI.
In late 2024, DeepSeek unveiled its next-generation large language model, V3, which was also open-sourcedThe performance metrics demonstrated that it outperformed several mainstream open-source models, while also maintaining a cost advantageDuring the World Economic Forum on January 20, 2025, DeepSeek turned heads once again with the launch of its latest open-source model, R1.
The R1 model introduced a significant breakthrough by utilizing pure deep learning methodologies to enable AI to exhibit spontaneous reasoning capabilities, competing effectively against OpenAI's flagship modelNotably, the training costs for the R1 model totaled only 5.6 million USD, significantly lower than the investments made by American giants like OpenAI and Google in the AI domain.
In Silicon Valley, DeepSeek has long been regarded as “the mysterious power from the East.” The release of the R1 model challenged the status quo in the AI industry, sharply affecting stock prices of tech companies in the U.S
- Yen Poised for Surge
- A-Share Market Confidence Boosted
- Historic Milestone in U.S. Stock Market
- Fed's Rate Hike: A Jolt to the Markets
- Short-term Growing Pains in the Semiconductor Industry
and climbing to the top of global download charts on platforms like Apple’s App Store and Google Play Store.
In essence, DeepSeek's approach has incorporated “smarter” algorithms that have reduced AI training costs by nearly 60% while simultaneously achieving or surpassing the performance of comparable modelsThis cost-performance ratio has resonated strongly within the industry.
Recognized as one of the most advanced large language models available, DeepSeek-R1 provides high-quality language processing capabilitiesLiu Wei, the director of the Human-Computer Interaction and Cognitive Engineering Laboratory at Beijing University of Posts and Telecommunications, highlighted that DeepSeek's greatest strength lies in its algorithmic improvements which save on computational power and data needs—illustrating a paradigm shift within the AI landscape.
Analysts believe that DeepSeek's innovations have disrupted the long-standing dependency on sheer computational power in the AI industry
This could unleash a wave of creativity and propel advancements in AI on a global scale, marking a departure from the traditional focus on heavy-lifting computational resources.
According to Huatai Securities, DeepSeek's combination of low costs and high performance has sparked discussions about capital expenditures and application scenarios, potentially prompting investors to reassess the technological potential of Chinese tech firms.
Zhejiang Business Securities echoed similar sentiments, suggesting that DeepSeek is stirring a global wave of innovation in AIThey also noted that the R1 model represents a successful challenge to overseas models, particularly to OpenAI, thereby democratizing AI capabilities and disrupting the established power structures based on computational strength.
Recently, multiple platforms including Baidu Intelligent Cloud, Huawei Cloud, Alibaba Cloud, Tencent Cloud, 360 Digital Security Company, and Yunzhuo Technology announced their integration of the DeepSeek large model
On the international front, leading tech firms such as Amazon AWS, Microsoft Azure, and Nvidia have also begun to engage with the DeepSeek model.
It is evident that the impact of DeepSeek on the AI sector is profound and far-reachingThe platform continues to shape the industry landscape, prompting fresh discussions about its potential commercial pathways.
On January 27, DeepSeek sparked a trading frenzy in the A-share market, witnessing several related concept stocks soarAs reported by Wind Data, the DeepSeek cooperation index surged by more than 13% at one point that day, closing the session with a 10.96% increase.
Among the notable gainers were Daily Interactive and Zhaochuang Information, which both closed with a ceiling increase of 20%. Other companies, including Torchat and Parallel Technology, also saw significant price surgesZhejiang Dongfang, an investor involved in the first round of financing for DeepSeek, experienced a 38.2% increase in share prices over the past ten days
Meanwhile, Huajin Capital, which participated in DeepSeek's Pre-A round, saw its shares rise by 21.7% within the same timeframe.
Many companies have capitalized on this wave of fortuneDaily Interactive announced on February 4 that it had integrated DeepSeek's model into its business operations and evaluated its applications, stating its commitment to embracing the open technological ecosystem offered by DeepSeek as the industry transitions to large models.
While some firms have sought to clarify their involvement with DeepSeek, investor excitement remains unabatedNumerous inquiries about the companies utilizing the DeepSeek model have flooded the platforms, demonstrating an ongoing interest in its potential.
Despite the hype, some “DeepSeek concept stocks” have shown signs of expected lossesDaily Interactive has indicated a forecast of net losses between 61.5 million to 91 million RMB for 2024. Yet, amid these challenges, the overall enthusiasm for DeepSeek’s transformative potential continues to grow.
However, DeepSeek's future in terms of commercialization remains a point of interest