AIAS is an EU-funded research project under the Horizon Europe Programme, specifically the Marie Skłodowska-Curie Actions (MSCA), aimed at strengthening the security and resilience of Artificial Intelligence (AI) systems against increasingly sophisticated cyber threats. The project focuses on developing an innovative AI-based cybersecurity platform that protects organisations’ AI-driven operations from adversarial attacks, including attempts to manipulate AI models, compromise their integrity or disrupt their functionality. By combining advanced AI defence techniques, intelligent deception mechanisms and explainable AI (XAI), AIAS addresses two complementary challenges: using AI to improve cybersecurity capabilities and protecting AI systems themselves from malicious attacks.

The AIAS platform integrates several innovative components, including an adversarial AI engine for simulating cyberattack scenarios, intelligent deception mechanisms based on honeypots, digital twins and virtual personas, and AI-driven tools for threat detection and mitigation. An explainable AI-based recommendation engine further supports cybersecurity professionals by providing transparent, actionable insights to inform security decisions and responses. With a particular focus on supporting SMEs, the project aims to deliver accessible and effective cybersecurity solutions that enhance organisations’ ability to anticipate, detect and respond to emerging threats. Through real-world validation, international research collaboration and knowledge exchange, AIAS contributes to advancing trustworthy AI, strengthening cybersecurity preparedness and improving the resilience of Europe’s digital ecosystem.