Photo: VCG

TheMK sport recent wide adoption of Chinese artificial intelligence (AI) startup DeepSeek reflects a market-driven choice, experts told the Global Times, emphasizing that Chinese companies’ technological progress should be viewed objectively. This comes after NASA, following the Pentagon’s move to block access to DeepSeek’s website, cited “security concerns” as justification and became the latest US federal agency to ban its employees from using the Chinese platform.

In a memo on Friday to all NASA personnel from the agency’s chief artificial intelligence officer, NASA employees were notified that DeepSeek’s servers “operate outside of the United States, raising national security and privacy concerns,” according to CNBC.

Prior to NASA, Bloomberg, citing a defense official, reported that US Defense Department employees connected their work computers to Chinese servers to access DeepSeek for at least two days before the Pentagon moved to shut off access. 

The Defense Information Systems Agency, which is responsible for the Pentagon’s IT networks, moved to block access to the Chinese startup’s website late Tuesday, the report said. 

Zhu Rongsheng, an expert from Tsinghua University's Center for International Security and Strategy, said the US response is a mix of surprise and skepticism – surprise that China has managed to train advanced AI models despite US export controls, and skepticism about the effectiveness of their own policies to restrict China’s tech development. 

For the US government, after deploying a wide array of sanctions, there is growing doubt about whether targeting China’s computing power is truly effective, alongside concerns about the US’ ability to maintain its leadership in AI development, Zhu said. 

Xiao Yanghua, a computer science professor at Fudan University, told the Global Times on Saturday that when China's technological advancements achieve market success and begin to challenge US counterparts, the US often moves to politicize the matter. 

This reflects a growing lack of confidence in the US as it faces repeated challenges from Chinese companies in the ongoing tech competition among major global powers, Xiao said. 

Echoing Xiao, Li Baiyang, an associate professor of intelligence studies with Nanjing University, told the Global Times on Saturday that the frustration, anger and powerless mindset among some US politicians on DeepSeek's breakthrough in terms of training costs and effectiveness is due to its struggling to adopt the right mindset toward China's advancements in science and technology.

Besides, Li said that hyping ideological and cross-border data risks, from another perspective, reveals an entrenched mindset in some Western governments: it has long been accustomed to controlling their domestic tech companies, leading to a presumption of guilt toward emerging foreign AI firms.

On Friday, the Netherlands said it will launch an investigation into DeepSeek's data collection practices and urged Dutch users to exercise caution with the company's software, Reuters said. 

Earlier, Italy's data protection authority announced Thursday it had blocked DeepSeek's AI services, citing insufficient transparency about how the company handles personal information. 

Despite that some countries are blocking the Chinese AI platform citing “security concerns,” many tech companies are integrating DeepSeek into their own platforms – a move widely interpreted by industry observers as a pragmatic acknowledgment of its technological strength. 

The GPU-maker Nvidia announced that DeepSeek-R1 now live with Nvidia NIM.  “DeepSeek-R1 is an open model with state-of-the-art reasoning capabilities. Instead of offering direct responses, reasoning models like DeepSeek-R1 perform multiple inference passes over a query, conducting chain-of-thought, consensus and search methods to generate the best answer,” Nvidia wrote on January 30.  

Nvidia stock dropped as much as 17 percent on Monday after the Chinese AI developer DeepSeek unveiled its DeepSeek-R1 LLM.

Nvidia's integration of DeepSeek occurred within a day of Microsoft launching the new R1 model on Azure AI Foundry and GitHub.

“DeepSeek R1 is accessible on a trusted, scalable, and enterprise-ready platform, enabling businesses to seamlessly integrate advanced AI while meeting SLAs, security, and responsible AI commitments,” Microsoft wrote in a statement. 

The stance of tech companies stands in stark contrast to the US government agencies' attempts to block DeepSeek. This indicates that the mainstream tech community in the US has a more objective and accurate understanding of DeepSeek's open-source approach, reflecting a recognition of its technical capabilities, Li told the Global Times. 

Tech media TechTarget wrote that the Chinese reasoning model has “caused a tsunami in the tech industry.” Cloud providers Amazon Web Services (AWS) and hardware maker Cerebras Systems have also engaged with DeepSeek. 

"We expect to see many more models like this – both large and small, proprietary and open source – excel at different tasks," AWS said in an emailed statement, TechTarget reported. 

Xiao stated that based on DeepSeek’s AI model, which is comparable to OpenAI’s, both domestic and international users can build on DeepSeek's open-source foundation and develop custom applications and enhancements. Such technological innovation has given DeepSeek a distinct commercial advantage.

“This is a reality the market must embrace, as its progress remains resilient despite political pressures,” Xiao said.