About the course
This course gives IEEE CertifAIEd professionals a working taxonomy for identifying, classifying, and mapping the ethical risks unique to generative AI and for connecting those risks back to the four IEEE CertifAIEd criteria.
Course Objectives:
The ethical risks introduced by generative AI do not fit neatly into frameworks built for discriminative AI. This microcredential equips IEEE CertifAIEd professionals to move from general concern to rigorous, taxonomical risk analysis.
Learners work through the IEEE 11016906 risk taxonomy and apply it to the failure modes practitioners actually encounter: training data toxicity, intellectual property exposure, hallucination, prompt injection, deepfakes, and the supply chain ambiguity between foundation model providers and downstream deployers. Every category is mapped back to the IEEE CertifAIEd Ethical Foundational Requirements, so learners leave with a working method for assessing generative AI systems against the framework they already hold a credential in.
Target audience
AI ethics officers, product managers, AI governance leads, and IEEE CertifAIEd Assessors working with or advising organizations that are deploying generative AI. Suitable for practitioners moving from foundational ethics work into specialist generative AI risk assessment.




Reviews
There are no reviews yet.