Research on Intelligent Traceability Framework for Trusted Artificial Intelligence |
|---|
| Qinmi Sun, Haibin Li |
- Abstract:
- With the deep integration of artificial intelligence and big data technology, the self-learning ability of the system brings efficiency improvement, but problems such as data pollution, algorithm black box, and model drift exacerbate the difficulty of tracing. This article proposes a three-layer traceability framework (TVB-Trace) that integrates blockchain metadata anchoring, dynamic verification mechanism, and trusted execution environment. By constructing a verifiable data lineage graph and algorithm decision chain throughout the entire lifecycle, it achieves transparent supervision of AI self-learning systems. Experiments have shown that this framework can improve data traceability accuracy to 99.2% and enhance model decision interpretability by over 40%. (Keywords: artificial intelligence traceability, blockchain, trusted computing, self-learning system).
- Download:
- IMEKO-TC8-11-24-2025-045.pdf
- DOI:
- 10.21014/tc8-2025.045
- Event details
- IMEKO TC:
- TC8
- Event name:
- IMEKO TC8, TC11 and TC24 Conference
- Title:
Joint conference of the TCs ‘Traceability in Metrology’ (IMEKO TC8), ‘Measurement in Testing, Inspection and Certification’ (IMEKO TC11), and ‘Chemical Measurements’ (IMEKO TC24).
- Place:
- Torino, ITALY
- Time:
- 14 September 2025 - 17 September 2025