Thesis Work - Generative AI for Scenario and Injection Design in IT/OT Cybersecurity Exercises

Description

Are you nearing the end of your studies and keen to apply what you know to work that supports the energy transition? You would write your thesis close to the business, alongside experienced specialists, on problems where your ideas can have a real effect.

The thesis sits within the Corporate Security & Resilience function at the company, a flexible, agile unit working in a demanding, fast-changing setting. The team builds security and resilience into all business operations and takes a holistic, risk-based approach to strengthening its security posture.

Cybersecurity exercises, tabletop and functional, help organisations stay resilient, and they matter most for critical infrastructure operators whose IT and OT environments are connected. How useful an exercise is depends on how realistic, technically accurate and relevant its scenario and injects (alerts, emails, media reports, decision points) are.

Building scenarios and injects by hand is slow and resource heavy. Planners must combine many variables into a sequence that feels credible to both IT and OT participants. Generative AI and large language models can help with scenario building, structured content and role-based dialogue, and agentic tools could propose follow-up injects so exercises adapt over time.

The thesis examines how generative AI can support both the design and the adaptive running of IT/OT cybersecurity exercises. It is a qualitative study that builds a conceptual framework and workflow specification for AI-assisted exercise design. Expert review and walkthroughs will test the framework for real-world use and relevance to IT/OT environments.

Expected outputs: an academic framework for generative AI-assisted scenario and inject design; a small library of reviewed example artifacts that demonstrate it; a documented workflow covering source preparation, scenario generation, state tracking, adaptive inject selection, human approval and debrief support; a limited demonstrator or simulation showing how participant decisions change later injects and outcomes; and a summary of expert findings on the risks and likely adoption problems they foresee.

The project is for 2 students.

Responsibilities

• Build the academic framework for generative AI-assisted scenario and inject design.
• Create a small set of reviewed example artifacts that show the framework in use.
• Document a workflow that covers source preparation, scenario generation, state tracking, adaptive inject selection, human approval and debrief support.
• Build a limited demonstrator or simulation in which participant choices alter later injects and outcomes.
• Gather expert opinions, analyse the risk factors and adoption problems they raise, and summarise them.

Requirements

• You are about to finish your academic studies and want to apply your knowledge in a real setting.
• You share the company's values: Active, Open, Positive and Safety.
• Your degree is in Computer Science or Software Engineering.
• You write and speak English fluently, since the thesis is written in English.
• The thesis is worth 30 credits at M.Sc level.

Conditions

Start date is 11 January 2027 or by agreement.

Location is Göteborg or Stockholm.

Application deadline is 18 November 2026.

Submit one file that combines your CV and academic transcripts, as only one upload is accepted.

Your contacts are Elpidoforos Arapantonis, who is both contact person and supervisor, at the company's site or +46 70 284 19 86.