HistoTrust : Tracing AI behavior with secure hardware and blockchain technology

Dylan Paulin (1, 2) , Raphaël Joud (1, 2) , Christine Hennebert (2, 1) , Pierre-Alain Moellic (2, 1) , Thibault Franco-Rondisson (2, 1) , Romain Jayles (2, 1)

1 CEA-LETI - Commissariat à l'énergie atomique et aux énergies alternatives - Laboratoire d'Electronique et de Technologie de l'Information
2 DSYS - Département Systèmes

In areas of activity where the notion of accountability is strong, the adoption of artificial intelligence (AI) is limited by the opacity and lack of understanding of its behavior, all the more so in the embedded domain where neural networks are compressed and executed on microcontrollers. While the NIST introduced in 2021 several principles allowing the AI explainability, this paper introduces a novel scheme, HistoTrust, combining secure hardware and blockchain technology to bring trust in the traceability of AI behavior and allow its explainability. HistoTrust attests in an Ethereum ledger all the relevant data produced by a physical device, especially the heuristics inferred by AI. Thus, the audition of the ledger allows security verifications and AI behavior analysis.

Domains: Engineering Sciences [physics]; Electronics; Computer Science [cs]; Hardware Architecture [cs.AR]; Computer Science [cs]; Cryptography and Security [cs.CR]; Computer Science [cs]; Neural and Evolutionary Computing [cs.NE]; Computer Science [cs]; Networking and Internet Architecture [cs.NI]

Dylan Paulin, Raphaël Joud, Christine Hennebert, Pierre-Alain Moellic, Thibault Franco-Rondisson, et al.. HistoTrust : Tracing AI behavior with secure hardware and blockchain technology. Annals of Telecommunications - annales des télécommunications, 2023, 2023, pp.15. ⟨10.1007/s12243-022-00943-6⟩. ⟨cea-03956052⟩