VeritasChain Releases VAP Scorecard Explorer v1.1, Improving Audit Defensibility Through Practitioner Feedback

Reference implementation update improves audit defensibility and scoring clarity based on practitioner feedback, without changing canonical benchmark criteria.

TOKYO, TOKYO, JAPAN, December 21, 2025 /EINPresswire.com/ -- VeritasChain today announced the release of VAP Scorecard Explorer v1.1, an update to the reference implementation used to apply the Verifiable AI Provenance (VAP) Benchmark in audit and regulatory contexts.

The v1.1 release does not modify the canonical VAP Benchmark criteria.
Instead, it focuses on improving scoring clarity, audit usability, and defensibility, based on practical feedback from a currently practicing Certified Public Accountant (CPA) working in audit and assurance, shared anonymously.

The VAP Scorecard Explorer is designed as an offline, implementation-agnostic assessment tool to support third-party evaluation of AI and algorithmic decision systems. It is not a product, certification, or endorsement, and it does not perform automated regulatory determinations.


■ Key Improvements in VAP Scorecard Explorer v1.1

The v1.1 update introduces several improvements aimed at aligning the tool more closely with real-world audit practice:

・Clear separation between evaluated and N/A criteria
Assessment results now distinguish between items that were evaluated and those that could not be assessed due to insufficient evidence, reducing misinterpretation of raw totals.

・Explicit “Score 2 requires” guidance
Each criterion now clearly explains what concrete evidence is required to achieve a full score, supporting transparent and defensible audit judgments.

・Outcome-based, implementation-neutral explanations
The Explorer emphasizes verifiable outcomes—such as cryptographic detectability of tampering or independent third-party verification—without mandating specific technologies or protocols.

・Improved consistency across UI, PDF, and JSON Evidence Packs
Exported assessment artifacts are now fully aligned, improving traceability and reviewer confidence.

・More transparent jurisdiction and regulatory profile alignment
Changes in jurisdiction or regulatory profile (e.g., EU AI Act vs. UK MiFID II) now result in clearly explainable scoring differences, reflecting real regulatory expectations.

・Research Context: From Logging to Verifiable Proof

The design direction behind v1.1 aligns with VeritasChain’s latest research paper,
“From Logging to Proof: Why Auditability Without Verifiability Fails Under the EU AI Act.”

The paper argues that logging alone is insufficient to meet the EU AI Act’s auditability and accountability requirements, and that verifiable, third-party-checkable evidence is necessary to support credible regulatory review.

The paper is publicly available with a DOI via Zenodo:

👉 https://doi.org/10.5281/zenodo.18000316

This research provides the conceptual foundation for the outcome-based, implementation-neutral approach reflected in the VAP Benchmark and Scorecard Explorer.


■ Designed for Defensible Assessments, Not Higher Scores

“The goal of v1.1 was not to raise scores,” said Tokachi Kamimura, Founder of VeritasChain and lead contributor to the VAP initiative.
“It was to make assessments clearer, more defensible, and easier to explain in real audit and regulatory discussions. The feedback we received from a practicing auditor was instrumental in shaping this update.”

The VAP Scorecard Explorer remains:

・Fully offline and self-contained
・A reference implementation, not a certification or regulatory approval
・Designed to support independent third-party judgment, not replace it


■ Availability

The VAP Benchmark (canonical criteria) and VAP Scorecard Explorer v1.1 are publicly available at:
👉 https://veritaschain.org/benchmark/
Feedback from auditors, regulators, and practitioners is welcome.


■ About VeritasChain

VeritasChain is developing open, implementation-agnostic infrastructure for verifiable audit trails and AI decision provenance, supporting regulatory requirements such as the EU AI Act, MiFID II, and other global audit and compliance frameworks. The VAP Benchmark and associated reference tools are designed to enable independent, third-party verification of AI and algorithmic systems without reliance on proprietary or vendor-specific technologies.


■ Media Contact
VeritasChain
Email: media@veritaschain.org
Website: https://veritaschain.org

TOKACHI KAMIMURA
VeritasChain Co., Ltd.
+ +81 7084849753
email us here
Visit us on social media:
LinkedIn
Facebook
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.