Trust4AI. Trustable AI-Driven Internet Search

The Trust4AI research project aims to explore new methods and tools for the automated detection of bugs in AI-driven search engines in accordance with EU trustworthiness compliance requirements. Trust4AI is part of the NGI Search project.

Publications

ASTRAL. A Tool for the Automated Safety Testing of Large Language Models

The ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA 2025).

Miriam Ugarte, Pablo Valle, Jose Antonio Parejo, Sergio Segura, Aitor Arrieta

Meta-Fair: Metamorphic Testing of Fairness in Large Language Models

Conference on Software Engineering and Databases (JISBD'25)

Miguel Romero-Arjona, José A. Parejo, Juan C. Alonso, Ana B. Sánchez, Aitor Arrieta, Sergio Segura

Red Teaming Contemporary AI Models: Insights from Spanish and Basque Perspectives

Conference on Software Engineering and Databases (JISBD'25)

Miguel Romero-Arjona, Pablo Valle, Juan C. Alonso, Ana B. Sánchez, Miriam Ugarte, Antonia Cazalilla, Vicente Cambrón, José A. Parejo, Aitor Arrieta, Sergio Segura

Testing the Evilness of Large Language Models

SBFT 2025 Search-Based and Fuzz testing (SBFT workshop at ICSE 2025).

Miguel Romero-Arjona, Aitor Arrieta

Early External Safety Testing of OpenAI's O3-Mini. Insights from Pre-Deployment Evaluation.

Technical Report

Aitor Arrieta, Miriam Ugarte, Pablo Valle, Jose Antonio Parejo, Sergio Segura

See all publications