Google Cloud · Generative AI Leader
Secure AI
Generative AI Leader · Lesson 13 of 14 · 1:49:58
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Prepare for the Google Cloud Generative AI Leader exam. In this lesson: security across the machine learning lifecycle (data poisoning, model source tampering, model exfiltration, prompt injection, sensitive data disclosure), Google's Secure AI Framework (SAIF) and its six core elements, and Google Cloud's security tools: secure-by-design infrastructure, Identity and Access Management, Security Command Center, workload monitoring, and Model Armor. Includes three original practice questions.
Preview the topics
- A shopper hides instructions in a chat message, to make Cymbal Retail’s assistant ignore its rules and reveal internal discounts. Which risk is this?
- Cymbal Retail’s security team wants the same security controls applied consistently, across every AI platform and every team in the company. Which SAIF element is this?
- Cymbal Retail’s chief information security officer wants one place to see misconfigurations, publicly exposed resources, and active threats across its Google Cloud projects, AI workloads included. Which tool fits?