Responsible AI Summit Workshop Day - Wednesday 23 September

AI governance frameworks were built for systems that assist human decision‑making. They are already being strained by agentic AI, and further capability leaps are coming faster than most organisations are prepared for. This workshop pushes beyond today's comfort zone, taking participants into structured, practical planning for the AI landscape of 2028–2029. Rather than asking if governance models will need to change, the session focuses on where they will fail first and how organisations can see those breaks coming.

Attendees will:

• Stress‑test their current AI governance frameworks against three realistic near‑future scenarios, including near‑AGI capability.

• Pinpoint the assumptions, controls, and processes most likely to break as AI capabilities jump again.

• Build early warning indicators tailored to their organisation, enabling earlier, more confident strategic intervention.

Sue Turner OBE

AI Governance Expert & Professor in Practice for AI and Technologies
University of Bristol Business School

See Bio

Responsible AI is no longer a technical or compliance issue alone — it is a leadership and trust challenge. Yet many Responsible AI initiatives struggle to gain traction at the executive level due to misaligned language, competing priorities, and unclear business value.

This interactive workshop is designed to help Responsible AI leaders, practitioners, and cross functional teams confidently engage senior leadership and the C suite on Responsible AI. The session focuses on how to communicate RAI in a way that resonates with business leaders, aligns with organisational strategy, and drives informed decision making without stalling innovation.

Through practical exercises and real world scenarios, participants will explore how to frame Responsible AI in terms of risk, value, reputation, and growth, and how to move conversations beyond principles to action.

Attendees will:

• Understand what C suite leaders care about when it comes to AI risk, opportunity, and trust, and how to tailor Responsible AI messaging accordingly.

• Translate Responsible AI concepts into clear, business relevant narratives that resonate with executive priorities.

• Build confidence in influencing senior stakeholders, handling challenge, and addressing common executive objections.

• Develop communication strategies to secure leadership buy in and align stakeholders across the organisation.

Walk away with practical frameworks, messaging tools, and leadership ready talking points to successfully position Responsible AI as a strategic enabler, balancing speed, innovation, and trust at the highest levels of the organisation.

Barbara Zapisetskaya

Principal Counsel
EBRD

See Bio

As AI systems gain unprecedented autonomy, traditional governance frameworks are struggling to keep up. Agentic AI is making decisions faster than humans can intervene, and organisations that cling to conventional oversight risk operational, ethical, and reputational failures.

This interactive workshop equips leaders with the tools and frameworks needed to design governance systems that are both robust and adaptable in the age of autonomous AI.

Attendees will:

• Reimagine decision rights and escalation paths for AI systems operating independently.

• Explore human-in-the-loop frameworks that scale without slowing innovation.

• Assign accountable technical leadership to ensure transparency and auditability.

Walk away with a practical blueprint for next-generation AI governance, including risk matrices, escalation protocols, and real-world dashboards to ensure your organisation can deploy autonomous systems safely, ethically, and at scale.

Andrew Harrison

Agentic & GenAI Governance Advisor
ABN AMRO Bank N.V.

See Bio

As AI agents become more autonomous and integrated into enterprise workflows, they introduce new and evolving attack surfaces, particularly through prompt injection and manipulation. Many organisations understand the concept of red teaming, but struggle to apply it effectively to dynamic, agent-based systems that interact with tools, data, and other models. For Security, Risk, and Responsible AI teams, testing these systems can feel complex, fast-moving, and difficult to operationalise.

This workshop takes a practical, hands-on approach to identifying and mitigating prompt injection risks in Agentic AI systems.

Attendees will:

• Understand how prompt injection attacks target AI agents and system workflows.

• Learn practical red teaming techniques for testing agentic systems.

• Explore how to design controls and guardrails against manipulation risks.

• Identify how to embed continuous testing into AI development and deployment.