When an AI system fails, the legal, operational and reputational consequences can escalate quickly, particularly when patient safety or sensitive data is involved. This scenario-driven LAB will take participants from the front end of AI contracting and vendor diligence through a simulated AI incident, challenging them to make the critical decisions that arise during the first 72 hours. Participants will leave with practical strategies for strengthening AI agreements, allocating risk and building an incident-response playbook before something goes wrong.

• Conducting AI vendor and model diligence
• Negotiating data rights, IP, warranties, indemnities and audit rights
• Addressing model changes, monitoring and cybersecurity
• Stress-testing a simulated AI failure involving inaccurate output, patient risk and data exposure
• Managing privilege, evidence preservation, regulator notification, vendor response and remediation