DeepSeek completely changed the tech world last month—and according to AI experts, for good reason. They say we might be witnessing the start of the influence of Chinese tech startups on the AI sector.
DeepSeek made headlines at the end of January with its R1 AI model, which the company claims can match the performance of OpenAI‘s O1 model at nearly the same cost. The fact that DeepSeek temporarily overtook ChatGPT to become the top app on Apple’s App Store caused a drop in tech stocks.
This achievement prompted American tech giants to question America’s position in the AI race against China and the billions of dollars behind those efforts. Although Vice President JD Vance did not mention DeepSeek or China by name in his speech at the Artificial Intelligence Action Summit in Paris on Tuesday, he certainly emphasized how big of a priority it is for the U.S. to lead in this area. He said, “The United States is leading in AI, and our administration plans to keep it that way,” while adding that “America wants to partner with other nations.” But it’s not just DeepSeek’s efficiency and power.
Experts say that DeepSeek’s R1 model, with its ability to reason and “think” through answers to provide high-quality results, as well as the company’s decision to make key parts of its technology publicly available, will advance the field.
Although AI has long been used in tech products, it has peaked in the past two years due to the rise of ChatGPT and other generative AI services, which have changed the way people work, communicate, and search for information. This has made chipmaker Nvidia a Wall Street favorite and flipped the path of Silicon Valley giants. Therefore, any development that can help create more capable and efficient models deserves close attention. Oren Etzioni, the former CEO of the Allen Institute for Artificial Intelligence, said, “This is not hype. But at the same time, this is a world that is moving very quickly.”
AI’s TikTok Moment
Tech leaders have immediately responded to the rise of DeepSeek. According to CNBC, Google DeepMind CEO Demis Hassabis called the hype around DeepSeek “exaggerated” but referred to its model as “perhaps the best work to come out of China.” Microsoft CEO Satya Nadella mentioned in the company’s quarterly earnings call in January that DeepSeek had some “real innovation,” while Apple CEO Tim Cook said during the iPhone maker’s earnings call that “innovation that increases efficiency is a good thing.”
But not all reactions have been positive. Semiconductor researcher SemiAnalysis has raised doubts about DeepSeek’s claim that it only costs $5.6 million to train. OpenAI told the Financial Times that they have evidence that DeepSeek used their models to train their competitor. An OpenAI spokesperson commented to CNN, “We are aware of this and are reviewing whether DeepSeek has improperly distilled our models, and as we gather more information, we will share our findings.” DeepSeek could not be reached for immediate comment.
Additionally, according to reports by the Associated Press and ABC News, after security researchers exposed potential links between DeepSeek and the Chinese government, a pair of U.S. lawmakers have already called for the app to be banned on government devices. Similar concerns have been raised about the popular social media app TikTok, which has been suggested to either be sold to an American owner or risk being banned in the U.S. “DeepSeek (large language models) is the TikTok of AI,” Etzioni said.
DeepSeek’s Profound Impact on the Tech World
Tech giants are already thinking about how DeepSeek’s technology might impact their products and services.
“What DeepSeek has given us is essentially a recipe in the form of a technical report, but they haven’t given us the additional missing parts,” said Louis Tunstall, a senior research scientist at Hugging Face, an AI platform that provides tools for developers.
Tunstall is leading Hugging Face’s effort to fully open-source DeepSeek’s R1 model; while DeepSeek has provided a research paper and model parameters, it has not disclosed the code or training data.
AI researchers, academics, and developers are still figuring out what DeepSeek’s advancements mean for the future of AI.
DeepSeek’s model is not only the first open-source model, nor is it the first model capable of reasoning through answers before providing them; last year’s OpenAI O1 model could do this too. DeepSeek is important because it can reason and learn from other models, and the fact that the AI community can see what’s happening behind the scenes. Users of the DeepSeek app utilizing the R1 model can also see its “thought” process while answering questions.
Durga Malladi, Senior Vice President and General Manager of Technology Planning and Edge Solutions at Qualcomm, told CNN, “You can see the wheels turning inside the machine.”
Tunstall believes that we may soon see a wave of new models that reason like DeepSeek. This could be significant because tech giants are racing to create AI agents, which Silicon Valley believes will be the next evolution of chatbots and how they interact with consumer devices—though this shift has not yet happened. On the social media platform X, the next version of the chatbot, Grok 3, will have “very powerful reasoning capabilities,” according to its owner, Elon Musk, in a video appearance at the World Government Summit on Thursday.
For now, the AI community will continue to play around with what DeepSeek has to offer—until the next breakthrough comes along.
