AI for good: the use of emerging technologies to address online threats
| Reference number | |
| Coordinator | Stockholms universitet - Inst för Data- och systemvetenskap |
| Funding from Vinnova | SEK 250 360 |
| Project duration | November 2024 - September 2025 |
| Status | Completed |
| Venture | International individual mobility within cutting-edge technology |
| Call | Closed offer - International individual mobility for cutting-edge technology 2024 |
Important results from the project
Within this project, we have developed and evaluated AI-based methods for detecting and assessing threatening communication in both Swedish and English. The work has deepened our understanding of model performance, potential biases, and the impact of training data on outcomes. The project has generated valuable practical insights for threat assessment and established a strong foundation for continued interdisciplinary research.
Expected long term effects
The project has provided several important synergies and opened up new opportunities for future collaborations and research directions. The project is now continuing through a collaboration with the FBI’s Behavioural Analysis Unit (BAU). This collaboration provides access to unique expertise in threat assessment, and contributes to the development and testing of our models. The collaboration will strengthen both the scientific quality and the practical applicability of the results.
Approach and implementation
The project began with digital work during the winter of 2024 and in June 2025 I traveled to the USA where I spent 9 weeks. During that time, the work was intensified through physical meetings that enabled more effective collaboration. The work followed the plan well and led to new ideas and collaborations. The collaboration worked excellent and the project created a strong foundation for continued interdisciplinary research.