How to read an AI regulation summary

News articles oversimplify — use this to extract **what actually applies to you**.

1. **Identify jurisdiction** — EU, US state, sector (health, finance, employment)

2. **Find the risk tier** — prohibited, high-risk, limited, minimal

3. **List obligations** — documentation, human review, bias testing, labeling

4. **Note effective dates** — grace periods vs enforcement

5. **Map to your system** — training data, deployment, users affected

6. **Read primary text** — summary + one official PDF beats hot takes

| Question | Why |

|----------|-----|

| Does it regulate **providers** or **deployers**? | Who must comply |

| Is **consent** required for biometric/likeness use? | Marketing & HR tools |

| Are **logs and audits** mandated? | Engineering work |

| What **penalties** exist? | Business risk |

Make a one-page internal memo: scope, gaps, owner, review date.

**Try the lessons:** `ai-regulation-basics` (Lane A) · `privacy-smart-sharing` (Lane A)