Identity (IAM) & cloud
Who is allowed to do what. This is how companies control AI tools.
Time this week: 6–8 hours
By Sunday you can explain authentication vs authorization, SSO, MFA and least privilege, and build a least-privilege access policy.
When a company rolls out ChatGPT Enterprise, Claude or Gemini, the first job is identity: SSO login, who gets access, and what admin rights exist. Most AI security controls start with IAM.
Start here: this week's labs
How SSO login to an AI tool works
Key ideas
Authentication (AuthN)
Proving who you are: password, phone code, fingerprint, security key.
Authorization (AuthZ)
What you are allowed to do after you log in: read, write, delete, admin.
MFA
Multi-factor authentication. Something you know + something you have (or are). Stops most password attacks.
SSO
Single sign-on. Log in once with your company account (Okta, Microsoft Entra ID, Google) to reach many apps, including AI tools.
Least privilege
Give each person or app only the access they need, nothing more. Very important for AI agents that can use tools.
Shared responsibility
In the cloud, the provider secures the building and hardware; you secure your accounts, data and settings.
Day-by-day plan
Tick each day when you finish it. Your ticks are saved in this browser.
Mini game: AuthN or AuthZ?
Practice outside this course (free)
The usual first security cert. A new version, V8 (SY0-801), launched on 17 Nov 2026. Check which version to book.Microsoft Learn: identity fundamentals (SC-900 path)
Free lessons on identity and cloud security.AWS: shared responsibility model
One page that explains who secures what.
These links need the internet. Everything else on this page works offline.
Exercises & worksheet
Weekly quiz
Mark it done to fill your progress bar on the course home.