A grounding in data poisoning as a deliberate, attacker driven manipulation of the data used to train or fine tune a machine learning model, distinct from an accidental data quality problem. You will learn the two broad categories, availability poisoning and targeted or backdoor poisoning, why a poisoned model can pass standard evaluation while still misbehaving under an attacker's chosen trigger, the specific exposure created by training on public or user contributed data, and why fine tuning on a smaller dataset represents a proportionally larger attack surface than large scale pretraining. The course covers data provenance and lineage tracking as the primary defense, validation and anomaly detection before training and their honest limits, and targeted post training testing for a suspected backdoor, closing with a worked hypothetical incident and a governance checklist for building a sustainable data integrity programme.
Nothing. Every course, exam, and certificate on the catalog is free — including retakes. All you need is a free Safeguard account.
None. The flagship course, Safeguard Certified Practitioner, is a beginner-level course — basic familiarity with how software is built helps, but every exam question is answerable from the lessons themselves.
You can retake it after a 24-hour cooldown, as many times as you need. Retakes are free, and each attempt draws a fresh random set of questions.
The Safeguard Certified Practitioner credential is valid for 24 months from issue. The expiry date is printed on the certificate and shown live on its public verification page. Renewing means passing the current exam again.
