DeepSeek, the Chinese artificial intelligence startup that sent tech stocks reeling this week, sparked fresh concerns about U.S. companies losing their competitive edge in the global race for AI dominance.
While DeepSeek claims to have built its R1 model with less than $6 million and limited computing power, some industry analysts say this development poses little threat to America’s AI leadership, with one even suggesting the technology might backfire against Chinese authorities as unwieldy AI could fuel information freedom and social dissent. The Chinese company’s approach, though efficient, comes as U.S. firms push ahead with more ambitious projects and infrastructure investments.
READ: OpenAI CEO Responds To DeepSeek’s R1, Emphasizes “More Compute” For Future AI
“If DeepSeek is as important as all of the papers say it is — and I think that there’s good reason to believe all of this — it makes OpenAI’s projects all the more valuable and all the more feasible in a shorter time period,” Will Rinehart, a senior technology researcher at the American Enterprise Institute, told the Daily Caller News Foundation. He also called the Chinese startup “the most important firm in AI right now” in his Monday article.
DeepSeek’s new R1 model, he explained, reduces both the financial barriers to AI development and reliance on advanced semiconductors — a key focus of the CHIPS Act. The model achieves comparable performance with significantly fewer parameters, requiring less computational power while maintaining output quality.
DeepSeek’s app has already cruised to the top spot on Apple’s App Store ranking, easily surpassing U.S. competitors like ChatGPT and Claude. The startup’s meteoric rise fueled investor doubts about Nvidia — whose hardware has long been considered essential for scaling advanced language models — and triggered a nearly 17% drop in the company’s stock by Monday’s market close. Silicon Valley titan Marc Andreessen even hailed the model as “AI’s Sputnik moment” — but there’s reason to doubt that DeepSeek’s R1 represents another instance of a communist power beating the U.S. to a major technological milestone.
READ: DeepSeek: The Rising Star In Open-Source AI Goes Down From Malicious Attack
Rinehart commended DeepSeek for pioneering high-quality AI at minimal cost compared to high-budget U.S. firms, but he doesn’t believe the Chinese startup’s success necessarily spells the downfall of U.S. firms like OpenAI. He explained that, because of its open-source structure, DeepSeek’s R1 offers OpenAI a kind of “playbook” with which it can learn to streamline its own models — opening the door for a symbiotic relationship between the two firms.
“What that means is, everyone else can look and see this white paper … and if you know what you’re looking at, you’re like, ‘OK, that’s what they’re doing.’ And so then you can basically avoid research dead ends and generally know which direction for the research to pursue,” Rinehart continued. “But as soon as that happens, that means that other relatively frontier labs can do the same thing.” The “white paper” Rinehart referenced is a document articulating a new language model feature and its applications for AI industry professionals, rather than consumers.
Although Rinehart believes that classifying software as simply “open-source” or “closed-source” is overly simplistic for today’s language models, they are still somewhat useful designations: an open-source model like DeepSeek’s is traditionally understood as software that is publicly available for anyone to use, modify or share, which promotes collaboration. A closed-source model like OpenAI’s is understood as one whose code is kept proprietary, limiting access to its underlying systems to authorized employees.
He also dismissed the notion that DeepSeek’s model is cost-efficient enough to eliminate OpenAI’s need for Stargate — the president’s $100 billion AI data center infrastructure initiative spearheaded by OpenAI CEO Sam Altman, Oracle CEO Larry Ellison and SoftBank CEO Masayoshi Son — to develop its “next-generation” of data centers to supply computing power for a much higher “projected usage” than currently demanded of DeepSeek.
“My sense is that this is going to continue regardless,” Rinehart said. “OpenAI is looking to basically create the next generation of data centers, and it seems like they’re going to do that regardless … what this new DeepSeek model suggests is that basically all of [OpenAI’s] efforts are potentially boosted at a much cheaper rate, right?”
“This shortens my timeline for AGI — for artificial general intelligence, which is what they’re specifically seeking … I mean, gotta give it to Trump, right? He announced one of the most major, important investments in data centers ever. And I think that’s still a win,” he added.
Artificial general intelligence (AGI) represents a key goal for the AI industry: an artificial intelligence system capable of matching or exceeding human-level performance across all cognitive or intellectual tasks — though AGI’s precise definition remains a subject of technical and philosophical debate.
Dan Schneider, a policy expert and vice president at the Media Research Center, a nonprofit watchdog group, seemed unfazed by the stock market’s reaction to the DeepSeek release in an interview with the DCNF. Nvidia, a longtime favorite among investors, lost nearly $600 billion in market value Monday, the largest single-day decline for any U.S. company in history.
“Nvidia’s stock prices had been soaring when people thought that it was a critical incumbent,” Schneider said. “OK, well, this is what happens in a free market system, where a new competitor enters the field and the incumbents then lose value. Wall Street should be optimistic about the value being created for all other businesses and people around the world. Whether you’re in agribusiness, or you’re in banking or you’re making widgets — AI is going to help create more value. This is going to be a great boon to the whole world economy.”
Schneider seemed similarly unconcerned about the Chinese company’s growing competitive edge against U.S. firms, arguing that DeepSeek poses less threat to American interests than to the Chinese Communist Party itself. He predicted the technology would eventually spark social upheaval in China, citing his belief that open-source “AI has a real opportunity to end all types of authoritarian and totalitarian regimes” through the free exchange of information.
“Ultimately, it’s consumers who are going to really benefit. I really believe that AI is going to be key to defeating all sorts of authoritarian and totalitarian regimes,” Schneider said, also acknowledging the Chinese Communist Party will likely “use any kind of technology at its disposal to create weapons” for “nefarious purposes.”
The Stargate Project has already begun preliminary construction for its first data center campus in Texas, according to OpenAI.
Please make a small donation to the Tampa Free Press to help sustain independent journalism. Your contribution enables us to continue delivering high-quality, local, and national news coverage.
Connect with us: Follow the Tampa Free Press on Facebook and Twitter for breaking news and updates.
Sign up: Subscribe to our free newsletter for a curated selection of top stories delivered straight to your inbox.
First published by the Daily Caller News Foundation.