Session 2: AI Governance After Deployment — Continuous Monitoring

Session 2: AI Governance After Deployment — Continuous Monitoring, Feedback & Iteration

Monday, 5th October | 12:00–13:15 | Online

Following our first session on AI Projects, Delivery & Governance, this session continues the conversation by looking at what happens after an AI solution is deployed into the real world.

Deployment is not the end of AI governance. Once an AI system encounters real users, changing data and changing business conditions, new evidence begins to emerge. Governance therefore needs to continue through monitoring, feedback, human oversight and iteration.

What We’ll Explore

We’ll discuss a project and governance perspective on areas such as:

• What continuous AI governance looks like after deployment
• Monitoring business and technical KPIs
• Data, model and context drift
• Human and domain validation
• Explainability and grounding
• Feedback loops and thresholds for intervention
• When to continue, investigate, adapt, retrain or reconsider an AI solution
• Keeping governance itself responsive as conditions change

I’ll share my perspective from an AI project management and governance lens, rather than presenting this as a deeply technical AI-development session.

If you work in AI development, data, security, governance, project delivery or another related area, please bring your own experience and perspective. The Q&A is intended to be a genuine peer discussion where we can share, challenge assumptions and learn from one another.

Session Format

Presentation: approximately 30–40 minutes
Open Q&A / discussion: approximately 30–40 minutes

If you missed our first AI Projects, Delivery & Governance session, you can watch the recording beforehand if you’d like. This session builds on that discussion, although you can still join without having attended Part 1.

🔗 Missed Session 1? Watch the YouTube recording here: https://youtu.be/_kcBRNXvbL8?is=Lftk-ey4Yl8KTd2H

Recording & Privacy

Only the presentation portion will be recorded.

The Q&A and open discussion will not be recorded.

Zoom privacy settings will be used so that participant cameras and names are not included in the presentation recording.

The aim of the Dublin AI Collaboration Space remains simple: share what we’re learning, hear different perspectives, and improve our understanding together.

Who Should Attend?

• Project Managers
• Scrum Masters
• Product Managers
• AI Practitioners
• Developers
• Business Analysts
• Governance Professionals
• Risk and Compliance Professionals
• Anyone interested in understanding AI Governance and Responsible AI
• Anyone curious to learn more about AI

From Session 1 to Session 2

Session 1 — 15 September:
BUSINESS → DATA → MODEL → DEPLOYMENT

Session 2 — 5th October:
DEPLOY → MONITOR → DETECT CHANGE → HUMAN/SYSTEM VALIDATION → GOVERN → INTERVENE/ITERATE → CONTINUE MONITORING

Looking forward to another good conversation.

Have a great day.

Thanks and Best Regards,
Shravanth