UK warns universities over China intelligence research links
A Warning That Names A Front
Every autumn, the doors of British universities open again. Lecture halls fill. Libraries reissue cards to returning students. Supervisors meet a new cohort of doctoral candidates. It repeats in stone quadrangles and in glass towers on the edge of industrial estates. It is also a market. Fees arrive from abroad. Partnerships are renewed. Research contracts are signed with institutions thousands of kilometres away. That ordinary rhythm reveals something simple: British higher education is woven into international networks it does not control. The threat came first. On Wednesday, MI5, the UK’s domestic security service, warned researchers and academics to cease work with the China Academy of General Technology. MI5 described the organisation as a “significant threat”.
The academy is also known as the China General Technology Research Institute. The accusation against it is specific. It is accused of paying UK academics working in sensitive fields, including artificial intelligence and cybersecurity, for research that serves an intelligence purpose. The MSS is China’s Ministry of State Security, the country’s intelligence agency. The claim, then, is not that collaboration happened. The claim is that collaboration was the cover. A partnership is normally assessed by what it produces: papers, trained students, shared data. The purpose alleged here is espionage.
More than 100 academics linked to the UK contributed to projects funded by China’s Ministry of State Security through that business that looks ordinary but works for another, hidden purpose. Three fields are named as areas of interest: artificial intelligence, cybersecurity and steganography. Steganography is the embedding of hidden data in pictures online. It is a technical discipline with civilian applications and obvious intelligence applications. China’s approach to intelligence gathering involves hoovering up vast amounts of data. The data often comes from people who do not know they are providing it. They believe they are taking part in normal research or normal consultancy work. So the warning is not only about individual misconduct.
The legal frame is domestic. The alert directs academic institutions and researchers to the 2023 National Security Act. That act makes it an offence to engage in conduct likely to help a foreign intelligence agency to a degree that counts under the law. The wording is broad. Its breadth is where the practical difficulty starts. A researcher must now judge, in advance, whether an ordinary collaboration could help an intelligence service to a degree that counts under the law. The warning also lands in a diplomatic moment. It risks casting a shadow over UK-China cooperation at a time when the bilateral relationship is enjoying a revival. Two facts sit awkwardly beside each other. One is a security service drawing a line. The other is a relationship warming up. Universities stand between them.
Following The Money Into The Lecture Hall
Chinese students are the largest group of non-EU international students in the UK, and their fees are a core line in many university budgets. For decades, institutions that are short of cash have turned to China for funding and collaboration opportunities. So a warning that lands on those institutions lands on their budgets as well as on their committees that approve research involving human or sensitive data. Refusing a partner is simple when the partner is marginal. It is harder when the partner is a line in the annual accounts. That is the trade at the centre of this case: security risk on one side, having enough money to pay what is owed on the other.
The scale of the academic relationship is documented. A transnational education partnership is not a vague memorandum. It is a structured programme that moves students, curricula and money across borders. That structure is why the warning reaches beyond a handful of laboratories.
Concern about the relationship is not new. The 2023 Intelligence and Security Committee report on China examined the academic link. It found that UK academic institutions “provide a rich feeding ground for China to achieve both political influence and economic advantage”. It also flagged the risk of China “directing or stealing UK academic research”. The two phrases describe two different mechanisms. Influence is a relationship. Theft is an event. A university can be exposed to both without noticing either. The committee framed academia as an environment rather than as a set of individuals. Environments are harder to police than individuals.
The warning is likely to send a chill through UK higher education. The institutions most exposed are those least able to absorb a lost contract. A university that declines Chinese funding must replace it or cut something. The alert names an organisation, not a boundary. It tells researchers to stop working with one entity. It leaves them to work out whether that entity was the only problem. The choice between a security risk that may never materialise and a financial risk that appears in the next budget is not resolved by the warning. It is handed to the institution instead. That is how a security alert becomes a management problem.
Where The Rules Stop And Risk Begins
Sam Goodman, the senior policy director at the China Strategic Risks Institute, describes UK academics as caught “between a rock and a hard place” when they navigate fields of advanced interest to Chinese intelligence. [1] The difficulty, in his account, is not the academics. It is the absence of instruction. His formulation: “There is a lack of a coherent steer from government on something like AI. It isn’t particularly clear on AI at this juncture what levels of collaboration we can have with the Chinese.” Artificial intelligence sits at the centre of the alert’s list of sensitive fields. It is also the field where the question of what counts as collaboration is hardest to answer. A rule that does not exist cannot be followed, checked or enforced. Caution becomes a local judgement rather than a national standard.
Charles Parton, a senior fellow at the Royal United Services Institute, sets a different priority. His starting point is the concession that “It’s all very well issuing these warnings”. The harder question, in his view, is technology by technology: “Actually far more important, surely, is a consideration of which technologies can be cooperated with China, and which can’t.” [1] Parton also doubts the value of naming companies. A list of prohibited firms would not help UK researchers perform due diligence on every potential partnership, because origin can be hidden. His reason is practical: “The Chinese are not stupid. If they want to hide the origin of someone, they’ll do it”. That is the case for categories over names. A list of names is something an intelligence service can plan around. A list of technologies, with their uses spelled out and their dual-use character explained, is something a university can actually plan with.
Meia Nouwens, a senior fellow at the International Institute for Strategic Studies, identifies a definitional gap: “A problem has been that in the past there hasn’t been a great deal of discussion of what areas of research constitute a national security risk.” [1] Her conclusion follows from that: “Without understanding the limits… of what the UK seeks to protect, it is impossible to have a conversation about managing risk.” A second point concerns ownership. “Research institutes that are conducting research on areas of strategic interest for China such as technology will have oversight from the party-state.” Party-state oversight is the norm for strategic research in China, not an exception. That does not mean every collaboration with a Chinese educational institution is inherently dangerous. It does mean that UK academics working in sensitive fields will feel increasingly uneasy about working with a country that has financially underpinned British higher education for decades. Both things hold at once.
This alert is not an isolated memo. This year, the UK’s National Cyber Security Centre warned businesses about Beijing-backed attempts to hack into everyday devices such as wifi routers. Last year, MI5 issued another China-related security alert, warning parliamentarians about approaches from two LinkedIn profiles thought to be linked to Chinese intelligence. Hardware, people and money: three surfaces that any open university sector touches. Set the coordinates. One axis runs from routers to social networks to funded research projects. The other runs from artificial intelligence to cybersecurity to steganography, the fields named as areas of interest. A collaboration sits somewhere on that grid, and universities are being asked to read the grid without a legend.
> Note: Only birds’ vision was tested; other predators may perceive patterns differently.