VeritasChain Shares Non-Normative Technical Note on EU AI Act PMM Evidence with European Commission and CEN-CENELEC
Non-binding, illustrative input on post-market monitoring evidence integrity under Article 72 of the EU AI Act
TOKYO, JAPAN, December 25, 2025 /EINPresswire.com/ -- VeritasChain Standards Organization (VSO), a non-profit and vendor-neutral technical standards body, today announced that it has shared a non-normative technical note through the official contact channel of the European Commission and directly with the CEN-CENELEC Joint Technical Committee 21 (JTC 21).The technical note provides an illustrative technical perspective on evidence integrity considerations for post-market monitoring (PMM) under Article 72 of the EU Artificial Intelligence Act (Regulation (EU) 2024/1689). The document is expressly non-normative and non-binding, and is shared solely as background technical material for informal discussion. It does not constitute a proposal, recommendation, regulatory interpretation, or request for endorsement or adoption.
Article 72 of the EU AI Act requires providers of high-risk AI systems to establish post-market monitoring systems that actively and systematically collect and analyze performance data throughout a system’s lifetime. While the Regulation defines PMM objectives, it deliberately leaves implementation details open. As a result, practical questions may arise regarding the robustness, integrity, and reliability of PMM evidence—particularly in post-incident investigations under Article 73 and in supervisory reviews.
In parallel, European supervisory authorities are sharpening expectations around AI-related operational resilience. For example, Germany’s BaFin recently issued non-binding guidance on the management of ICT risks associated with AI use, emphasizing risk identification, monitoring, and mitigation across the AI lifecycle. This supervisory direction is consistent with broader EU governance trends, including the EU AI Act and the Digital Operational Resilience Act (DORA).
VSO’s technical note explores these questions at a high level, focusing on challenges such as long-term evidence integrity, traceability across system updates, post-incident reconstruction, and provider-deployer accountability boundaries. It outlines tamper-evident logging mechanisms as one possible technical approach among several, emphasizing that multiple valid implementation pathways may exist and that minimum PMM compliance does not require any specific technical architecture.
The note was made publicly available to ensure transparency and accessibility for technical stakeholders, including regulators, market surveillance authorities, and standards experts. The document can be accessed at:
https://github.com/veritaschain/vso-docs/blob/main/VSO-TECH-NOTE-PMM-001-CENELEC.pdf
“By sharing this material as a non-normative, illustrative technical input, we aim to contribute to constructive technical dialogue around PMM implementation without advocating any particular solution,” said Tokachi Kamimura, representing VeritasChain Standards Organization. “Our intention is to support informed discussion by outlining practical considerations that may be relevant as PMM practices and harmonized standards evolve.”
VeritasChain Standards Organization has consistently engaged with international public bodies by providing non-normative technical inputs for information and dialogue purposes only. This approach reflects VSO’s commitment to neutrality, transparency, and respect for regulatory and standardization processes.
TOKACHI KAMIMURA
VeritasChain Co., Ltd.
+ +81 70-8484-9753
email us here
Visit us on social media:
LinkedIn
Facebook
YouTube
X
Other
Legal Disclaimer:
EIN Presswire provides this news content "as is" without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
