Content Co-Pilot Regulated content

Strict

JR

Risk level

Status

Team

Channel

Guardrails

AI assists drafting; humans approve. Every action — flags, edits, approvals — is logged for audit.

Pipeline

Content queue

All items currently in review.

Title	Type	Channel	Team	Status	Risk	Updated

No matching content

Try clearing filters or adjusting your search.

Compose

New content draft

Provide a brief or paste rough copy. The co-pilot drafts, evaluates risk, and routes for review.

Setup

What you're creating and for whom

Content type

Channel

Audience

Owning team

Content

Brief and draft

Campaign or use-case context (optional)

Draft content

First drafts start in Needs Review.

Risk preview

Live as you type

How risk scoring works

Transparent, rule-based — every flag is explainable.

The co-pilot scans each draft against a library of rule patterns derived from healthcare-content policy. Each match contributes a weighted score; the total maps to a risk band:

Low · 0–24 Medium · 25–54 High · 55+

What's flagged: guaranteed outcomes, medical overreach, diagnosis-like phrasing, fear-based messaging, misleading certainty, privacy-sensitive wording, unauthorized medical advice, missing disclaimers, and unsupported clinical claims.

Strict mode tightens thresholds (Medium becomes 15+, High becomes 40+) for sensitive campaigns or audiences.

Every flag includes the matched phrase, the policy reason, and a safer-rewrite suggestion. Reviewers and Compliance Leads can accept rewrites or escalate. Nothing publishes without explicit human approval.

Why this was flagged

1 of 7

Version comparison