Privacy, secrets and trust boundaries
Anything you paste may leave your trust boundary; minimize and redact.
What you will be able to do
- Never paste API keys, tokens, private keys or full customer records.
- Redact names, emails and identifiers before sharing code.
- Treat tool-using agents as code execution: sandbox and limit permissions.
In plain English
Prompts can contain personal data, credentials or proprietary code. Use approved tools, remove secrets, and check data retention settings. Never let a model execute destructive commands without a human review step.
# Check for common secrets before sharing a snippet
grep -nE 'API_KEY|SECRET|TOKEN|PRIVATE KEY|password' ./snippet.txt
# Replace real identities with stable placeholders
sed -E 's/[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+/user@example.com/g' snippet.txtYour turn
EasyRedact a snippet
List three categories of data to remove before sharing code with an AI assistant.
TEXT · starter
1.Check your understanding before moving on.
Knowledge checkQuestion 1
Which data should never be pasted into an unapproved AI tool?
Select an answer to continue
Study tools
Make it yours.
Keep a private note for this lesson, then turn the key ideas into flashcards you can revisit.
Sign in to save notes and flashcards.Both are private to your account and stay with your progress.
Sign inFinished reading?
Lock in what you just learned.
0%
Mark this lesson complete to save your place and move one step closer to the certificate.