Launching with OC-AISF and OC-AIRT · Team programmes scoped on enquiryODA3 Institute ↗
Sign inEnquire
OC-MIDPASD · AI Security: DefensivePractitionerIn development

Model Integrity & Data Poisoning Defense

In development · enrolment sequence to be confirmed

LevelPractitioner
PrerequisitesFoundation or equivalent
StatusIn development
SpecificationDraft · not approved
EnrolmentNot yet open

01Overview

Model provenance verification, poisoning detection and integrity validation controls.

Domain scope · ASD

The defensive core: AI threat taxonomy, control-plane implementation, token security, pipeline and model integrity, and AI incident response.

Intended for: Security practitioners moving into AI security or deepening it

02Proposed case references

Source review pending
I-2
OpenAI/DeepSeek Model Distillation CampaignSource review · In review

Proposed connection, subject to source review. I-2 is the anchor case. Model provenance and integrity controls are taught against its documented distillation pattern.

03Assessment

Format, duration, passing score and practical component are published with the approved specification.

04Instructor

TBA
Announced when appointed

Instructors are practitioners appointed against a written standard for the ASD domain.

How faculty are appointed →

Official specification registerOC-MIDPSpecification not yet issued
Version Last reviewed
This course is in active development. Duration, pricing, session dates and instructor are published when confirmed. Proposed case references, framework references, syllabus, labs and assessment are drafts until the course specification is approved.