Governance
What "governed AI workflows" actually means in practice
Governed is an overused word. Here is what it should mean when an AI agent is doing real work: boundaries, decision points, a record, access control and limits.
Governance
Designing human approval steps so people stay in charge
Where to place approvals in an AI workflow, what the approver should see, how to avoid rubber-stamping, and how to handle rejection and silence.
Governance
Human-in-the-loop AI: when an agent should ask before it acts
Four situations where an AI agent should stop and ask a person first, what the person should see, and how to keep the queue short enough to stay careful.
Governance
EU AI Act Article 14: human oversight in plain English
What Article 14 of the EU AI Act asks of human oversight, who it applies to, the new 2027 dates and how workflow builders can design for it. Not legal advice.
Governance
The agent audit trail: what to log so you can defend a decision
What an AI agent audit trail should record, how long to keep it, how to handle personal data in logs, and how retries and re-runs fit in.
Governance
An AI governance policy template for automation teams (one page, copy it)
A one-page AI governance policy for teams that build workflows and agents: scope, roles, approval tiers, logging and review. Copy it and edit.
Governance
Prompt injection in workflow agents: a defender's checklist
How hostile text in emails, documents and web pages can steer an AI agent, and a practical checklist of limits, gates and tests that reduce the damage.
No posts match that search. Try a shorter term, or clear the filters.
Get new posts by email
Practical notes on governed AI workflows, in your inbox. Unsubscribe whenever you like.
Build your first governed workflow today.
Start on the free plan, bring your own model key, and see a real run traced end to end.