📙 AI as an Accountable Entity: Governing Risk When Machines Make Decisions ⚖️
summary
AI decisions become risk-bearing services that need inventory, a simple materiality score (impact, autonomy, sensitivity) and assigned owners across model, business and control teams; a loan model scores 3/3/3 and therefore requires governance. Practical controls include business sign-off, ethics review, quarterly bias testing, retrain-as-deploy processes, structured explainability logs per decision and continuous assurance with semi‑annual audits. Accountability rests on answering what happened, why, who owns it and providing reproducible evidence over time.
AI-written summary · about
chapters (9)
- 0:00Introduction and welcome
- 1:06Why AI governance matters
- 3:38AI decisions vs traditional software
- 8:17AI as a risk-bearing system
- 12:52Accountability triangle for AI
- 16:25Governance controls and oversight
- 18:50Auditing AI systems
- 20:34Redistributing accountability
- 22:00Q&A session
related talks
31mAI risk governance
30mResponsible and Ethical Governance of AI Models
52mThe Risky Business, of AI Illiteracy
08mWhy AI Governance Matters More Than Ever
22mA story arc for responsible AI
39mBuilding a Responsible AI Governance Program
24m2025: year of Ai dilemma - ethics, regulations and innovations
30mAgentic AI in Action: From Framework to Deployment