Privacy law reaches AI at every stage: when training data is collected, when prompts and logs are stored, when a model may itself contain personal data, and when outputs drive decisions about people. Regulators in the EU and US have started to say how existing law applies, and new rules are layering on top. This course teaches you to analyze each stage under the three regimes most organizations face.
Under the GDPR you will work through EDPB Opinion 28/2024 on AI models, the legitimate-interest test for training and deployment, Article 22 and the right to an explanation of automated decisions after CJEU Case C-203/22 (Dun & Bradstreet Austria), DPIAs and their interplay with the EU AI Act fundamental rights impact assessment, data subject rights against trained models, international transfers, and enforcement actions by data protection authorities. You will separate what has been adopted (the 2026 AI Omnibus amending the AI Act) from the Digital Omnibus GDPR amendments that remain under negotiation. In the US you will apply the CCPA regulations on automated decisionmaking technology, risk assessments and cybersecurity audits, other state privacy laws with profiling opt-outs, and sensitive-data statutes such as Illinois BIPA and Washington's My Health My Data Act. For health AI you will cover business associate agreements, de-identification, ambient scribing, the Section 1557 rule on patient care decision support tools, and the pending HIPAA Security Rule update. Educational, not legal advice; the Wirewalk Institute certificate attests to completion and assessment of this course only.