The Chinese AI start-up DeepSeek has evidently shook the tech world last week with its low-cost AI assistant, which is reportedly 20 to 50 times cheaper to train and operate than OpenAI’s models. The announcement sent shockwaves through the stock market, wiping nearly $600 billion from Nvidia’s market value as investors feared reduced demand for its processors.
However, Pat Gelsinger, the former CEO of Intel and a computing industry veteran, believes the market got it wrong. In a LinkedIn post brimming with optimism (and just a touch of defiance), Gelsinger argued that DeepSeek’s breakthrough is actually good news for the AI industry—and even better news for companies like Nvidia.
Gelsinger’s lessons from DeepSeek
In his post, Gelsinger shared three key takeaways from DeepSeek’s rise, each grounded in decades of lessons from the computing world:1. Lower costs expand the market Gelsinger dismissed fears that cheaper AI models would hurt the demand for computing power. Instead, he compared the situation to the rise of PCs and mobile devices, where making technology affordable triggered explosive adoption.
“The market reaction is wrong. Lowering the cost of AI will expand the market,” Gelsinger wrote. “Today I am an Nvidia and AI stock buyer and happy to benefit from lower prices.”
Gelsinger argued that affordable AI will lead to broader integration into applications, growing the overall pie for everyone involved.2. Constraints fuel innovation DeepSeek’s success is a testament to creativity under pressure. Faced with export restrictions and limited hardware resources, the startup managed to create a world-class AI solution. Gelsinger noted that many of the greatest breakthroughs in computing were born under significant constraints.3. Openness drives progress Gelsinger also called out the shift toward proprietary AI systems, saying it stifles innovation and transparency. He credited DeepSeek for embracing an open-source approach, which encourages collaboration and scrutiny.
“Open wins every time it is given a proper shot,” Gelsinger wrote. “AI is much too important for our future to allow a closed ecosystem to ever emerge as the one and only in this space.”
Gelsinger isn’t alone in questioning the market’s reaction. While DeepSeek’s AI model can run on local devices like PCs and smartphones, training it still requires advanced data centre-grade hardware—a market dominated by Nvidia, AMD, and Broadcom.
“The real money in AI is providing the chips for the data centres,” said Daniel Morgan, senior portfolio manager at Synovus Trust Company, in an interview with Reuters. “Overall, I view the AI tech selloff today as an opportunity to add high-quality tech shares on weakness.”
Even as companies like DeepSeek optimise their models to use fewer resources, the demand for high-end processors is likely to grow as firms like OpenAI and Anthropic scale up their models with more parameters, requiring massive computing power.
DeepSeek’s breakthrough has put a spotlight on cost-efficient AI and shaken up the status quo dominated by big players like Nvidia, OpenAI, and Microsoft. But as Gelsinger pointed out, the implications aren’t as simple as they might seem.
While the startup’s innovation opens the door to wider AI adoption, it doesn’t diminish the need for advanced processors in large-scale data centres. The balance between affordability and cutting-edge performance will likely shape the future of the AI hardware market.
For Gelsinger, the lesson is clear: don’t panic. Instead, see this as an opportunity to innovate, expand, and embrace the spirit of openness. As he cheekily put it: “I’m buying Nvidia stocks—and I’m happy about it.”