Global Workspace Theory

AI Agent Governance: Evidence Is Not Authority

Published by

TL;DR AI agent governance must separate evidence, approval, and permission to execute. A shared workspace can coordinate a convincing recommendation without granting any component the authority to act. Preserve provenance through summaries, bind approvals to exact operations and resources, and recheck authorization when the action reaches execution. Keep production credentials outside the reasoning components, account … Explore: AI Agent Governance: Evidence Is Not Authority

Designing a Shared Workspace for AI Agents

Published by

TL;DR A shared workspace for AI agents should be a bounded, versioned view of an active task, not an expanding conversation transcript. Specialists contribute observations, interpretations, and proposals through controlled interfaces. The workspace preserves their differences, tracks dependencies, and makes conflicting evidence visible. Separate active state from recovery history and reusable memory. Use atomic updates … Explore: Designing a Shared Workspace for AI Agents