HiPEAC

Making certifiable AI a reality for critical systems: Core demo

As fully autonomous systems become increasingly crucial for advanced functions, AI takes on an increasingly dominant role in critical systems. However, current AI software is developed as a black-box that is hard to verify, lacks clear requirements, and clashes with the rigorous, certifiable and explainable processes required by safety-critical systems.

The SAFEXPLAIN project is working to close this gap through its next- generation open software platform designed to show how AI can meet the stringent requirements of functional safety in autonomous technologies.

The SAFEXPLAIN Core Demo brings a simplified yet comprehensive instantiation of SAFEXPLAIN technologies and tools, together with the platform’s key resources, on a reference AI-based system. It will be used as a concrete reference to shows our end-to-end approach to making certifiable AI a reality for critical systems by ensuring that systems are safe by construction a follow the latest relevant standards.

Making certifiable AI a reality for critical systems: Core demo
Part 4 of our webinar series
Discover how SAFEXPLAIN technology can accommodate scenarios with critical functionalities in three selected toy examples from the automotive, rail and space domains. 
18 June 2025 | 10h | HiPEAC Webinars

Program

Carlo Donzella has 40 years of experience in R&D and innovation management, project management, and quality assurance in the digital technology sector, with applications across multiple industries, especially in mobile, transportation, automotive, and media. He is the co- founder and CEO of exida Development SRL, established in 2008, a company part of the internationally renowned exida Group in the field of functional safety. He is also the co-founder and CEO of exida Engineering SRL/GmbH and other affiliated entities within the exida Group. As a reviewer for dozens of international programs, he has served in various administrative and advisory roles for many public institutions and private companies. He is ASPICE® Principal Assessor (HWE, MLE, Cybersecurity extensions), CFSE™, CSP™, Board Member of SPIN Italia, Team Leader in SAFEXPLAIN and EdgeAI-Trust EU Horizon projects.

Dr. Enrico Mezzetti has been an Established Researcher at the Barcelona Supercomputing Center since 2016. He holds a PhD in Computer Science from University of Bologna (Italy) on software timing analysis and industrial development process. He authored more than 60 publications on the verification of critical embedded systems, specializing in timing characterization and analysis of multicore systems, RTOS, and software architectures.


Metadata

Application areas: Automotive, Avionics, Space

Topics: Artificial intelligence, Deep learning, Dependability, IoT, Safety


Summary

Explore the SAFEXPLAIN Core Demo, showcasing how AI can meet stringent safety standards in critical systems. Learn about its application in automotive, rail, and space domains for safety by design.