Engineering Dependable AI Systems

Morayo Adedjouma (1) , Christophe Alix (2) , Loic Cantat (3) , Eric Jenn (4) , Juliette Mattioli (2) , Boris Robert (4, 2) , Fabien Tschirhart (3) , Jean-Luc Voirin (5)

1 DILS (CEA, LIST) - Département Ingénierie Logiciels et Systèmes
2 THALES [France]
3 IRT SystemX
4 IRT Saint Exupéry - Institut de Recherche Technologique
5 Thales DMS France, SAS

If AI algorithms are now pervasive in our daily life, they essentially deliver non-critical services, i.e., services which failures remain socially and economically acceptable. In order to introduce those algorithms in critical systems, new engineering practices must be defined to give a justified trust in the capability of the system to deliver the intended services. In this paper, we give an overview of the approach that we have put in place to reach this goal in the framework of the French Confiance.ai program. Based on the needs of the industrial partners of the program, we propose a model-based analysis framework capturing the two dimensions of the problem: the one related to the development and operation of the system and the one related to the trust in the system.

Mots clés: Systems engineering and theory; System verification; Mission critical systems; Artificial intelligence

Domaines: Intelligence artificielle [cs.AI]; Ingénierie, finance et science [cs.CE]

Morayo Adedjouma, Christophe Alix, Loic Cantat, Eric Jenn, Juliette Mattioli, et al.. Engineering Dependable AI Systems. 17th Annual System of Systems Engineering Conference (SOSE), IEEE, Jun 2022, Rochester, United States. ⟨hal-03700300⟩