The increasing integration of artificial intelligence (AI) into sectors such as critical infrastructure, autonomous transport systems, industrial automation, and digital services provides immense transformative potential. However, this integration also introduces significant vulnerabilities, as AI systems are susceptible to new forms of cyber threats and operational failures. Verifiable trustworthy and secure mechanisms to ensure the resilience of AI systems against such threats are thus necessary, especially in times of geopolitical instability. To address this, RESIST systematically builds a world-class multi-disciplinary research center, bringing together leading researchers and national initiatives for driving research and development of cyber-resilient AI systems for the benefit of Swedish industry and society.
RESIST - Resilience and Security for Trustworthy AI Systems
Our vision is to make Sweden a role model in secure trustworthy AI by pioneering cyber resilience across the AI lifecycle.
Trustworthy AI requires strong cyber resilience
As AI becomes rapidly woven into the critical functions of society, the need to understand how to make these technologies resilient, in an uncertain and rapidly changing world, is growing just as quickly.
Strong roots in AI and cybersecurity
The overarching scientific challenge is to develop novel techniques and methods with strong roots in AI and cybersecurity to achieve holistic cyber resilience throughout the AI lifecycle. This means that even during the face of adversarial conditions and unforeseen disturbances maintain:
- security
- privacy
- operational integrity
- trustworthiness for AI models, systems and agents
The two communities, AI and security, cover complementary aspects of AI security but tend to rarely intersect. Security researchers bring domain rigor and practical threat models, while AI researchers emphasize robustness and new methodologies. The result is a field that is growing in volume but divided in vision, leaving significant opportunities unexplored at the intersection of the two. RESIST will focus on this intersection.
Linköping University
hosts RESIST. The funding of SEK 60 million comes from the Swedish Foundation for Strategic Research SSF.
About RESIST
RESIST consists of
Research program
RESIST develops methods for:
- ensuring the trustworthiness and verifiability of AI models during their entire lifecycle
- securing AI models against runtime attacks
- ensuring that systems developed using AI are robust and secure
- establishing resilient AI architectures that can withstand failures without compromising performance.
These four themes are the starting point of the RESIST research program. As the field is very dynamic and rapidly evolving, we will track progress and adapt the themes accordingly.
Validation program
- AI-driven critical infrastructure
- AI-empowered software services
- AI-automated robots and vehicles
To ensure that the methods and solutions developed by the center can benefit society, the center runs an extensive program for evaluating, testing, and validating its results in collaboration with external stakeholders.
Activities
- Strategic research agenda
- Education and training
- Policy and strategy advise
- Internationalization
- Dissemination
RESIST functions as a national hub for expertise in cyber‑resilient AI, contributing to new courses and improved education, disseminating knowledge through events and public lectures, and supporting policymakers in the development of effective policies.
To ensure effective operations and international excellence, the center has
a management
consisting of a director, a manager, and deputy director,
a board
and
an international
scientific advisory committee
Contact
Researchers at Linköping University
Researchers at Lund University
Researchers at RISE
- Shahid Raza
- Apostolos Pyrgelis
Researchers at Örebro University
Collaboration for impact
RESIST consists of researchers
within AI‑ and cybersecurity from Linköping University, Lund University, RISE and Örebro University,
and builds on the strengths of
two WASP cybersecurity NESTs, two WASP assistant professors, one WASP guest professor, and several related projects and national initiatives.
The consortium has very strong
international connections especially in the EU and is supported by leading companies and organizations such as Saab, Ericsson and FOI among others.
