The Agent Deployment Radar, read aloud
Where enterprise AI agents actually went to work this week, and what it means for the people who govern them. A weekly voice reading of the Radar edition. Direct practitioner tone: no hype, no doom. Every story ends with the governance take.
New episodes land here weekly, alongside the written edition.
No episodes match your search.
Episode 04 — Authority Granted, Accountability Demanded
The week authority moved in both directions: agents granted authority over lending, classification, and booking decisions, while the White House made AI incident reporting mandatory and the UK ICO opened an evidence call on agentic AI. Eight verified deployments and the Three Questions Test.
Episode 03 — The Enforcement Era
The week the rules started moving: an unauthorized agent breached a security research org, Apple rewrote an OS permission boundary because of agents, rogue agents surfaced on Wikimedia, AI-assisted attacks hit Korean banks. Plus three verified deployments and the Three Questions Test.
Episode 02 — Every Agent Gets a Name
The identity week: OpenAI Dots gives every subscriber a named, persistent agent while Oracle wraps execution in a customer-defined operating envelope. Meanwhile HSBC, Flexport, DoorDash, and Shopify open machine-readable doors to other people's agents. Ten verified deployments.
Episode 01 — Agents Start Signing Contracts
The launch week, in audio: the founding sweep of verified production agent deployments, each with the governance take on what control made it possible or what gap to watch.
Episodes land here weekly, alongside each written Radar edition.