Agenda

AI governance frameworks, policies, treaties, and multilateral agreements, are routinely developed without being tested under the conditions in which they must actually function.  Tabletop exercises rehearse operational decision chains.  No equivalent methodology exists for the governance layer itself.  The result is a structural gap: frameworks are assumed to work until a grey zone/"under the threshold" crisis proves otherwise.

Auracelle AI Governance Labs, in partnership with Bath Spa University, has developed a methodology to close this gap: asynchronous stress-tested governance wargaming.  Rather than testing who responds and how fast, it tests whether the governance framework itself, the policy, the treaty clause, the regulatory provision, holds under adversarial pressure, coalition fracture, and real-time adversarial shock.

WSIS+20 arrives at a moment when AI is already shaping escalation dynamics, disinformation environments, logistics networks, and institutional trust, faster than governance frameworks can adapt.  The vision this session advances is straightforward: governance frameworks should stress-test.  Simulation is the missing institutional layer.  It provides the evidence base for policymakers, treaty negotiators, and standards bodies to make informed decisions about the resilience of the rules they are building.