Research
Research and development
We turn research into technology and capability that work in the field.
We research how people and AI work together, how to use external information and expert knowledge, and technology for analysis and validation from an attacker's perspective. The results grow into methods, tools and foundations for collaboration, and improve through real work and feedback from the field.
What do we research? Reviewed
Problems from the field become research, and the results are checked in the field again.
Five core areas
The core capabilities behind every service. What we learn in the field comes back into the next round of research.
- COLLABPeople & AI Collaboration
Technology for people and AI to share context and results while keeping their own tools and ways of working.
- INTELExposure Intelligence
Technology that collects and analyzes external observations, relationships and history to support investigation, monitoring and judgment.
- KNOWHOWKnowledge Infrastructure
A foundation that links expert knowledge and field experience to their sources and context, for people and many kinds of AI to use.
- EXPLOITOffensive Engineering
Technology and tools for research, analysis and security validation from an attacker's perspective.
- AGENTAgentic AI
Execution and evaluation technology that lets AI plan, use tools and carry out work.
The five areas have no order and shape one another. COLLAB and AGENT research also reaches beyond security.
Research topics and environments
Where the core capabilities are put to work: the problems we study and the environments they come from.
- AI, LLM and agent security
The attack surface of AI models and agent services, and responses to prompt attacks, data leakage and misuse.
- IT systems security
Attacking and defending web, mobile, APIs, cloud and networks: the fundamentals every other topic builds on.
- OT and physical systems
From OT, ICS, IoT and firmware to robots, vehicles and ships: vulnerabilities in systems that touch the physical world, and their effect on safety.
- Advanced offense and defense
Offensive and defensive techniques against advanced threats, including zero-day vulnerability research.
- People and AI working as a team
How several people and several AI agents work effectively as one team, and how to shape that into workflows governed by roles, permissions, approval and verification.
Environments we work in
From the field, back to the field
Research starts from problems met in real work. The results become methods, tools and foundations for collaboration, and are tested again in the field.
In the services we deliver
Assessment, validation and investigation become faster and more consistent. You receive the result as better-grounded work, not as a separate product.
In training, and for more organizations
The same methods support hands-on training and help more organizations reach a higher standard of security.