This fireside chat will cover:
- Should banks govern GenAI as a one-off intake or continuously across design, deployment, monitoring, and change, especially for agentic systems?
- How can first-line teams adopt model-risk practices such as pragmatic testing, immutable prompt and action logs, operational dashboards, and rapid kill switches without stifling innovation?
- What does SR262 mean for GenAI and agent tech; does it sit outside traditional model-risk frameworks, and how conservatively should firms interpret and document controls?