Cybercampus Sveriges enables cybersafety and cyber defence research beyond what is possible for a single university, institute, government agency or company. Current projects, responsible researchers and any results are presented here.
When Anonymized Data Isn't Private: How Large Language Models Reveal Emerging Weak Points
Alejandro Russo
Researcher
Alejandro Russo
Description
Classic anonymisation was built for smaller, slower data worlds. Today’s massive, linkable datasets—and off-the-shelf LLMs—make privacy attacks easier by lowering skill barriers and automating expert-style reasoning. Resilience now depends on data minimisation, prompt controls, and formal guarantees like differential privacy to counter uncontrolled linkage.
LLM-Powered Vulnerability Discovery and Analysis in Open-Source Software
Reseachers
This project is conducted by Syafiq Al Atiiq and Christian Gehrmann at
Vypr AI AB
, a start-up company from Lund University.
Description
Is it possible to discover vulnerabilities in open-source software more quickly? Syafiq Al Atiiq and Christian Gehrmann demonstrate AI capabilities in cybersecurity by discovering real vulnerabilities in critical open-source infrastructure. Scanning major open-source projects to identify and responsibly disclose security vulnerabilities.
Sweden’s Cybersecurity at Crossroads
Researchers
The project is led by Henrik Glimstedt, includes postdoc Annika Andreasson (PhD from KTH CDIS), and is co-funded by the Stockholm School of Economics.
Description
This research provides a long-term, interdisciplinary platform to support strategic policy learning, measure the implementation of the national cybersecurity strategy, and aims to generate actionable insights by combining historical institutionalism, comparative policy analysis, case studies, stakeholder interviews, an international panel of experts, and scenario-based activities/exercises, to shape Sweden’s cyber resilience towards 2040.
Readiness to teach cyberhygiene – an exploration of the perceptions of primary school teachers (ReaTCH)
Researchers
Cybercampus co-funds this project at Jönköping School of Engineering and Kodcentrum, and it is led by Joakim Kävrestad and Erik Bergström.
Description
As more educational initiatives at different levels of society have been developed, it becomes crucial that we have methods to evaluate their effects. The project aims to initiate the development of methods for evaluation of learning materials for teaching cyberhygiene in primary school by conducting interviews with teachers and other educators.
Machine Learning over Encrypted Data
Researcher
PhD student Anton Israelsson
Description
Using homomorphic encryption (FHE), it is possible to perform computations directly on encrypted data. This enables processing and analysis without ever exposing the underlying information. Potentially, this technology can allow researchers to analyse confidential data from non-trusting parties, such as hospitals, without ever sacrificing data ownership. The aim of this project is to improve the scalability and efficiency of multi-party FHE and explore machine learning on encrypted data.
