UAIF GOVERNANCE

UAIF Defensive Publication Template

Searchable HTML edition of UAIF-DPR-007, with the approved source retained as the downloadable record.

STATUSFinal aligned public template
ACCESSPublic
FRAMEWORKUAIF v1.0
SOURCEUAIF-DPR-007

Document download

Internal review, authorization, submission and archive workflow is excluded.

Document status and notice

ODA3 Institute

ODA3 Pvt Ltd

UAIF Defensive Publication Record

Template for Algorithm Prior Art Disclosure

Document ReferenceVersionStatusDateClassification
UAIF-DPR-007v1.0TEMPLATEQ2 2026PUBLIC
Published ByGoverning LawContact
ODA3 Pvt Ltd | ODA3 InstituteIndia (default)CONTACT_AT_ODA3_DOT_ORG
PURPOSE OF THIS TEMPLATE This template is used by ODA3 Institute (ODA3 Pvt Ltd) to place UAIF algorithms into the public prior art record. This defensive publication programme is part of ODA3 Institute’s intellectual property policy under the GAISSF Ecosystem Licence (GEL v1.0), ensuring conformant implementations of UAIF algorithms are free from patent encumbrance. Defensive publication prevents any party — including ODA3 — from obtaining patents that could block conformant implementations of UAIF. Each algorithm subject to this policy is published separately using this template and submitted to IP.com or an equivalent database. Questions: CONTACT_AT_ODA3_DOT_ORG | IP Policy: Supporting Document D1.

PART A: Publication Record Metadata

FieldValue
Publication Title[ALGORITHM NAME — e.g. 'UAIF Diminishing-Contribution Context Modifier Algorithm']
Publication Reference[e.g. ODA3-DP-001]
Publication Date[DATE — this is the prior art date]
Authors / Inventors[Names of individuals who developed the algorithm]
AssigneeODA3 Pvt Ltd (ODA3 Institute)
Framework Version[e.g. UAIF v1.0]
Classification (IPC)[International Patent Classification code(s) — e.g. G06F 40/30]
Database Submission[IP.com / Derwent / other — include submission confirmation number once obtained]
ODA3 Non-AssertionCovered by ODA3 Patent Non-Assertion Covenant (UAIF IP Policy, Supporting Document D1, Section 4.1) and UAIF Technical Specification Section 13.3 under GEL v1.0. (Note: Where Supporting Document D1 conflicts with GEL v1.0 or TCR §13.3, GEL v1.0 and TCR §13.3 prevail.)

PART B: Technical Disclosure

This section must be completed in sufficient technical detail to establish prior art — i.e., to enable a person skilled in the art to implement the algorithm from this disclosure alone. The standard is the same as for patent specification enablement. Incomplete disclosures do not establish robust prior art.

B.1 Technical Field

[Describe the technical field of the algorithm — e.g. 'This disclosure relates to multi-factor severity scoring systems for AI incident classification, specifically to methods for combining multiple contextual modifier values while preserving analytical differentiation and preventing score saturation.']

B.2 Background

[Describe the problem the algorithm solves and the prior art context — e.g. existing approaches and their limitations. This contextualises the algorithm and demonstrates novelty.]

Example for UAIF Context Modifier Algorithm:

Example Background Prior AI incident severity scoring systems (including CVSS-derived approaches) apply contextual modifiers by simple addition or multiplication, leading to score saturation when multiple contextual factors are present. Saturation obscures analytical differentiation between compound and simple attacks. A method that preserves all contextual information in the score while preventing saturation was needed.

B.3 Detailed Description of the Algorithm

[Provide complete, step-by-step description of the algorithm including: inputs, processing steps, decision points, outputs, and edge case handling. Include pseudocode or reference implementation code. The description must be sufficient to implement the algorithm without additional information.]

Example structure for a UAIF algorithm disclosure:

ElementRequired Content
InputsList and type all inputs with valid ranges and constraints
Processing stepsNumbered sequential steps; each step must be unambiguous
Mathematical operationsExpress all formulas in standard notation; define all variables
Decision logicDocument all conditional branches and their conditions
OutputsList and type all outputs with precision requirements
Edge casesDocument behaviour for empty inputs, boundary values, invalid inputs
PseudocodeProvide language-agnostic pseudocode implementing the full algorithm
Reference implementationReference to the GEL v1.0-licensed reference implementation (e.g. UAIF v1.0 reference code, available at https://docs.oda3.org/uaif/conformance)

B.4 Claims Scope (Informal)

[Describe in plain language what aspects of the algorithm are being placed in the prior art record. This is NOT a patent claim — it is a plain language description of the inventive concept being defensively published. Example:]

Example Claims Scope The following inventive concepts are placed in the prior art record by this publication: 1. A method of combining multiple contextual severity modifiers using a diminishing contribution factor, where each successive modifier contributes a fixed fraction (e.g. 50%) of the prior modifier's contribution, combined with a per-factor cap and a total sum cap. 2. The application of such a method to AI incident severity scoring, where the full set of applied modifiers and their individual contributions are recorded in a machine-readable audit trail regardless of their contribution magnitude.

B.5 Variations and Equivalents

[Describe obvious variations of the algorithm that should also be captured in the prior art record — e.g. different diminishing factors (33% instead of 50%), different cap structures, application to different scoring domains. This broadens the prior art and prevents narrow workarounds from being patented.]

PART C: Non-Assertion Statement

ODA3 Patent Non-Assertion — ODA3 Pvt Ltd ODA3 Pvt Ltd (trading as ODA3 Institute) irrevocably commits that it will not file patent applications on the algorithm(s) described in Part B of this Publication Record, and that it will not assert any patent claims it owns or controls that read on the algorithm(s) described herein against any party implementing those algorithms in conformance with the UAIF specification as published by ODA3 Institute under the GAISSF Ecosystem Licence (GEL v1.0). This commitment is consistent with the ODA3 Patent Non-Assertion Covenant set out in the UAIF Intellectual Property Policy (Supporting Document D1, Section 4) and the Patent Non-Assertion Covenant in the UAIF Technical Specification Section 13.3 under GEL v1.0. Where Supporting Document D1 conflicts with GEL v1.0 or TCR §13.3, GEL v1.0 and TCR §13.3 prevail. Version boundaries (MAJOR, MINOR, PATCH) are determined solely by ODA3 Institute in accordance with GEL v1.0 Section 10.2. This commitment is irrevocable with respect to the UAIF MAJOR version current at the date of this publication and all subsequent MINOR and PATCH versions within that MAJOR version. This non-assertion commitment does not grant any right to use ODA3 Institute’s trademarks, including GAISSF™, UAIF®, AI-IRF®, and ODA3®, in connection with any implementation. Such use requires separate written authorisation from ODA3 Institute, consistent with GEL v1.0 Section 9.3. A party implementing a defensively published algorithm may make factual alignment statements (e.g. “implements UAIF context modifier algorithm as described in ODA3-DP-XXX”) without ODA3 certification, consistent with GEL v1.0 Section 10.3. Use of UAIF Certified or UAIF-Compatible marks still requires written authorisation from ODA3 Institute.

Public template boundary

Parts A–C are published to explain the metadata, enabling technical disclosure and non-assertion record expected for a defensive publication. Internal technical/legal review, authorised signatory fields, repository selection, submission credentials, archive operations and approval workflow are withheld.

Template only

Completing this page does not publish prior art, create an ODA3 non-assertion commitment or authorize submission. A record becomes effective only through an approved publication and archive process.