Updating the interface and loading the latest catalog signals.
Adversarial testing is a method for systematically evaluating an ML model with the intent of learning how it behaves when provided with malicious or inadvertently harmful input. This guide describes an example adversarial testing workflow for generative AI.
Google ML and GenAI learning across Cloud and developer education.
Mar 28, 2026
Sign in to bookmark this course, mark it in progress, and save completion notes or evidence.
Sign in to trackFlag courses that are broken, irrelevant, duplicated, or clearly stale. Reports are intentionally short so review can happen quickly.
Google ML and GenAI learning across Cloud and developer education. Source: Google catalog connector.
The primary action takes you to the provider-owned destination.
Recency remains visible so users can weigh trust and timing quickly.
Signed-in users can move directly from evaluation into tracking behavior.