A production agent deleted the database in nine seconds. The board’s next question is upstream of the model.
Agentic AI governance demands a pre-deployment accountability check. The PocketOS incident shows why boards must ask before agents go live.
Category: AI Governance
You’re hiring a Chief AI Officer. What exactly are you hiring them to do?
The APS Chief AI Officer mandate creates a role by June 30. Whether that role has the authority and information access it needs to lead AI strategy is a question most agencies will not have asked before the appointment is finalised.
December 2026: the automated decision deadline most boards haven’t planned for
The December 2026 Privacy Act disclosure obligation sits on top of an accountability architecture most organisations haven’t built. Disclosing what you cannot defend is worse than not disclosing.
Sovereign AI is being decided in your architecture choices, not your AI policy
Architecture meetings are locking Australia’s sovereign AI compliance posture now. The SOCI Act and March 2026 data centre expectations create governance obligations that cannot wait for policy to catch up.
Stay up to date:
ยฉ 2019 All rights reserved