Frequently Asked Questions About Cybersecurity

The higher education sector is particularly vulnerable to cyber-attack. This is due to the large amount of personal and research data that universities – including library systems – routinely store. Beyond the direct cost of an attack, state-sponsored espionage can damage the value of research (notably in STEM subjects), reduce public and private investment in affected universities, and erode national knowledge benefits.  

It is not always easy to identify the perpetrators of a breach, but two main groups are typically involved: criminals seeking data for financial gain, and state-sponsored actors seeking personal data or intellectual property for strategic advantage.  

Pirate sites in academic publishing are online platforms that provide unauthorized, free access to stolen, copyrighted research papers, journal articles, and books. They include Sci-Hub, Library Genesis (LibGen), Z-Library, and Anna’s Archive. These platforms not only disrupt the business models of legitimate publishers, they also threaten the integrity of academic research. 

Sci-Hub is one of several known pirate sites. It hosts research papers stolen from publisher platforms, often using stolen credentials. The US Justice Department has previously investigated reported links between Sci-Hub's founder and Russian intelligence. This means that using Sci-Hub to access research could carry consequences well beyond getting access to illegally downloaded content.

Pirate sites like Sci-Hub have the potential to damage the research process itself. Unlike publishers, they have no incentive to ensure the accuracy of research articles, uphold ethical standards, or correct and retract content where issues arise. Publishers – working alongside bodies like the Committee on Publication Ethics (COPE) – take seriously their duty of care for the scholarly record.

They correct inaccuracies, plagiarism, or falsified data when they occur. They also invest in the infrastructure (such as Crossref) that keeps research properly linked and verifiable. Sites like Sci-Hub undermine the scholarly communications framework as they operate by taking what others have invested in. This comes at a time when publishers are working to make research more open and accessible by legitimate means.  

Unfortunately, yes. Large language models (LLMs) – the technology behind AI tools such as ChatGPT and Claude – are trained on vast amounts of text data. There is growing evidence that some AI developers have used pirate sites, including Sci-Hub and LibGen, as training data sources, without the consent of publishers or authors. This raises serious concerns. Not only does it represent large-scale copyright infringement, but it may also mean that AI systems are being built on research content that has been retracted or not properly verified and corrected as needed.

This means there is no guarantee of accuracy or integrity in the foundation of these tools. SNSI monitors these developments closely as part of its broader work on threats to the scholarly record. 

There is no single, easy answer to this question. We must work together towards a shared goal: to safeguard a successful, secure research experience. To do this, we need to ensure legitimate institutional and individual access to high-quality, licensed, peer-reviewed publications, protect data, and safeguard the entitlements of licensed institutions.