Surveys that reveal risk
Design surveys able to detect the emotional factors linked to a person's vulnerability to the different social engineering techniques.
Europa Se Siente
Hard2bit research and development project co-funded by the European Union through the European Regional Development Fund (ERDF) and managed by CDTI.
Many cyberattacks start with a person: an email impersonating a supplier, a call that creates urgency, a message that appeals to trust. Social engineering exploits emotions, and the usual measures treat every employee the same way.
The project develops a prototype tool that builds cybersecurity risk profiles for the people in an organization. It starts from the emotions detected in answers to psychological surveys, which an artificial intelligence model classifies together with their context. Once the level of risk is determined, the organization can apply preventive measures tailored to each case and continuously improve its protection.
Design surveys able to detect the emotional factors linked to a person's vulnerability to the different social engineering techniques.
Select the language representation models best suited to interpreting the meaning and emotional weight of the answers.
Train and tune two deep learning networks: one for semantic analysis and one for emotional analysis.
Build a working prototype with a web dashboard that shows the type and level of risk and the associated preventive measures.
| Title | Deep learning networks as a preventive technology against social hacking (official title: «Redes DL como tecnología preventiva del hacking social») |
|---|---|
| File number | IDI-20240645 (EXP 00167166) |
| Beneficiary | Hard2bit, S.L. |
| Programme | CDTI R&D Projects, co-funded by the European Regional Development Fund (ERDF) under the Multiregional Programme for Spain 2021-2027 |
| Thematic scope | Cybersecurity and artificial intelligence |
| Geographical scope | Madrid region, Spain |
| Duration | April 2024 to December 2026 |
| Project budget | €713,664.00 |
| European Union support | €606,614.40 |
The approved work plan is organized in four phases. The first project milestone was certified by CDTI in July 2026.
Initial architecture of the two neural networks and first selection of language representations.
Survey design and first version of both networks.
Parameter definition and first semantic-emotional analyses.
Tool with a web dashboard, validated with real data.
This section is updated with the results achieved at each milestone. The final results will be published when the project closes, planned for December 2026.
The project is carried out in collaboration with the University of Granada, through its Department of Computer Science and Artificial Intelligence, which contributes its experience in designing deep learning techniques for text analysis.
Last updated:
Before you leave…
Quick 15-minute assessment and we'll tell you what to prioritise first: Microsoft 365, pentesting, vulnerability management, SOC, DORA, NIS2, ENS or ISO 27001.
No spam. Reply within 24h.