On Monday, Jan. 27th, the Chinese hedge fund HIgh-Flyers unveiled its latest Large Language Model (LLM): DeepSeek, an open source competitor to ChatGPT.
The release of Deepseek has sparked concern from federal government regulators and Silicon Valley tech firms over the ease of its development in spite of U.S. sanctions on GPU chips.
“The release of Deepseek AI from a Chinese company should be a wake up call for our industries that we need to be laser focused on competing to win — because we have the greatest scientists in the world,” said U.S. President Donald Trump in a press conference on Jan. 28th.
Shortly after this speech, he met up with Nvidia CEO Jensen Huang to discuss the ramifications of DeepSeek and what limits on exports, if any, should be implemented.
The AI sector has been grappling with soaring chip prices as companies scramble to secure high-performance GPUs that are essential for training sophisticated AI models. Nvidia’s dominance in the GPU market has contributed to rising costs, pushing AI companies to explore alternative solutions. DeepSeek’s entry into the market disrupts the notion that the best state of the art quality chips are required to produce competitive advanced LLM AI models.
This success comes in spite of U.S. sanctions restricting China’s access to high-powered GPU chips. The U.S. has imposed export controls on advanced semiconductor technology to limit China’s ability to develop cutting-edge AI applications. However, DeepSeek has managed to work around these restrictions by leveraging domestically developed GPU chips and optimizing its training processes to reduce dependence on high-end U.S. hardware.
DeepSeek’s primary advantage lies in its proprietary model architecture and its ability to operate with significantly lower computational costs. Unlike other LLMs, which require massive datasets and expensive hardware to develop, DeepSeek has leveraged optimized algorithms and innovative data-curation techniques to train its models more efficiently. This breakthrough allows for reduced dependency on high-end GPUs, making advanced AI systems more accessible to a broader range of developers and organizations.
The new LLM reportedly demonstrates superior contextual reasoning and adaptability, setting it apart from existing AI models. With capabilities that rival OpenAI’s GPT-4 and Google’s Gemini, DeepSeek’s potential impact on industries ranging from software development to customer service and content generation is substantial.
Following DeepSeek’s announcement, AI-related stocks have experienced noticeable fluctuations. Shares of Nvidia, which has benefited immensely from the AI boom, saw a 17% dip on Monday, Feb. 24th as investors speculated on the potential for reduced GPU demand.
Regulatory bodies and government officials have taken notice of DeepSeek’s advancements. The U.S. and European Union have been actively working on AI governance policies, and DeepSeek’s open source framework could pose a threat to our financial institutions, particularly in the technological sector. When Nvidia shares dipped, it represented a total loss of $465 billion.
Regulations in response to deepseek’s debut have already started impacting the Queens College community. On Feb. 21st, QC’s Chief Information Officer Troy Hahn sent out an email stating:
“Effective immediately, installing or using DeepSeek on any Queens College devices, servers, or networks is strictly prohibited. This policy applies to all faculty, staff, and students utilizing college-owned equipment or accessing our institutional networks.”
This directive was implemented to comply with NY Governor Kathy Hochul’s issuance of a statewide ban on Deepseek on devices or networks.
“Public safety is my top priority and we’re working aggressively to protect New Yorkers from foreign and domestic threats,” said Governor Hochul at a Feb. 10th press conference.
