British policymakers and technology experts are urging the government to reject the growing influx of low-cost artificial intelligence tools from China, citing significant risks to national security and data privacy. The warning comes as Chinese developers increasingly offer open-weight models that allow Western businesses to deploy advanced AI without paying expensive licensing fees.
While these tools are attractive for their affordability, critics argue they carry hidden costs related to political censorship and state surveillance. The debate mirrors broader concerns about Chinese electric vehicles in the UK, where consumers often overlook how Beijing subsidizes exports to undermine foreign industries.
DeepSeek and the censorship risk
At the center of the concern is DeepSeek, a prominent Chinese AI developer that releases new versions of its models monthly. These open-weight systems allow users to download and run the software on their own servers, a feature that some analysts claim adds variety to the global AI landscape dominated by US firms like OpenAI and Anthropic.
However, Ross Burley, director of the Centre for Information Resilience, argues that this approach does not eliminate risks. He notes that running a model locally does not remove political censorship embedded in the software or repair underlying security weaknesses. An evaluation by the US National Institute of Standards and Technology found evidence of political censorship in downloaded DeepSeek models, even when hosted on independent infrastructure.
The policy for DeepSeek’s online chatbot states that it collects prompts and personal data, processes them in China, and permits disclosure to Chinese authorities. This raises questions about the integrity of systems used by UK businesses and public services for critical decision-making.
Data privacy and state surveillance
Michael Wooldridge, a professor of AI at the University of Oxford, stated that it is reasonable to assume data inputted into these chatbots could be shared with the Chinese state. This potential for data exfiltration poses a profound threat to UK policymaking and commercial confidentiality.
The concern extends beyond data privacy to the broader geopolitical strategy of Beijing. China is using its economic leverage to create dependencies, similar to its Belt and Road Initiative, which turns states into clients reliant on Chinese technology. Unlike Russia, which uses AI for cyber-attacks to create chaos, China aims to widen its circle of influence by offering goods without raising awkward questions.
Experts warn that relying on Chinese technology, from electric cars to AI, will hollow out domestic industries and undermine the security of British data. The UK is advised to seek better regulation of US AI while supporting domestic and European suppliers to ensure a more secure and benign technological future.
Source: The Guardian
