STALE: OUR MEASUREMENT

Our published benchmark result was measured with v0.41.2, 5 releases ago, past this project's own 2-release window, and the newest real-model measurement, dated 2026-09-18, is past this project's own 7-day window. That is our own number, not your assessment, which runs against your policy on the current release. Why, and what unblocks it

STALE: REGULATORY CLOCK

The regulatory entries on this page were last checked against their primary sources on — 34 days before this build’s anchor of , past this project’s own 30-day window. Every date here was true when it was read and has not been re-read since. Verify against the primary source before relying on it. The clock, with every source

ProvaelGitHub, star the repositoryQuickstart
ISO/IEC 23894:2023

ISO/IEC 23894:2023

ISO/IEC 23894:2023 applies ISO 31000 risk-management practice to AI, and its hard part is naming the risks that apply to your system. Provael contributes to that step: a named risk list, the Embodied AI Security Top 10, and a measured rate per risk as input to the risk assessment.

Timing Published 6 February 2023. Guidance on managing AI risk - it is not certifiable, and it does not tell you which risks to carry.

Use it for

What a Provael report can be used for

Risk identification & assessment

In plain words: a taxonomy gives the risk register its rows, and a measured rate per row turns a qualitative likelihood into something with a denominator. Provael’s catalogue entry reads “AI risk management — risk identification & assessment input”, with the signal “The EAI taxonomy as the mapped AI-risk context and the measured rate per risk as risk-assessment input to the AI risk-management process”.

The evidence

  • The Embodied AI Security Top 10 as a ready risk context for a physical-AI system, so the register starts from a named set rather than a blank page.
  • A measured rate per risk, with its interval and benign control, as risk-assessment input - evidence toward likelihood, not a determination of it.
  • Floor results carried alongside: risks that were tested and did not separate from the benign control are reported at their own counts, which is what stops a register recording only the risks that happened to fire.
Limits

What it cannot be used for

A mapping identifies possible relevance. It does not establish that this framework applies to your product, that anything here is compliance, that an assessor or insurer will accept it, or a real-world risk probability. Whether it applies to your product depends on your role, market and product class: a question for your counsel, not a determination this page makes.

  • Simulation evidence for the tested checkpoint and protocol. Physical safety and conformity require separate assessment.
  • Provael measures adversarial robustness - susceptibility to manipulation - not general accuracy, reliability, or functional safety.
  • The measured result uses templated, auditable attacks; the search-based families (optimized, universal_patch, gradient_patch) have no real-policy rate. Results are a floor on susceptibility - a behavioural lower bound, not a certified worst-case bound.
  • Provael’s own Embodied AI Security Top 10 is an independent taxonomy maintained by Provael, not one of the instruments on this page; a mapping onto it is a categorisation, not a legal or standards requirement.
Evidence, not certification

Running Provael does not make a system compliant or certified - it generates measurements you can put into a conformity or assurance file. Provael is not a notified body, a lab, or a certification scheme.

Independent project. Not affiliated with or endorsed by ISO, the EU, NIST, IEC, OWASP, or MITRE. Not legal advice.

Clause references are indicative: check each against the text you hold before you cite it.

What it is

The framework

  • Guidance for organisations that develop, produce, deploy or use AI, on managing AI-specific risk and integrating it into existing risk processes.
  • It is built on ISO 31000, so it describes a process - identify, analyse, evaluate, treat - rather than a checklist of controls.
  • It is guidance, not a certifiable standard. ISO/IEC 42001:2023 is the certifiable management-system counterpart.
  • It does not supply a domain risk catalogue. For a physical-AI system you have to bring one, which is the gap the Embodied AI Security Top 10 is written to fill.
Timing

Dates (each line shows when it was checked)

Published
6 February 2023ISO/IEC JTC 1/SC 42 (checked 10 August 2026)

Check the live EUR-Lex or ISO text before relying on a date.

Sources

Primary references

Sample SARIF ↓All frameworks

See what a report contains.

The sample report comes from a public run and is published in full, nothing redacted. To have this mapping filled in for your own policy, request a scoping call.

See the sample reportRequest a scoping call