In regulated medicine, every promotional sentence is a compliance event — and one slip is a regulatory event. One claim that drifts past the label, one missed adverse event, one fair-balance line left off, and a routine conversation becomes a matter for the regulator. So teams hold back: fewer touches, tighter scripts, narrower coverage. Reach is what loses.
Zenreps resolves that tension the other way. The discipline is built into how a zenrep speaks — and the proof is not a promise, it is a record. On the gated text channels, web chat and Telegram, every reply passes the compliance gate before a single word reaches the clinician, and the gate writes a receipt while it works.
One question, four checks
Take one exchange. An HCP writes in: "What's the once-daily dose of Veltarin, and does food matter?" Before the answer is delivered, the gate runs its checks: a scope check (in scope — an on-label dosing question), grounding (matched to an approved labeling passage), fair balance (the required safety information paired automatically), and an audit record (question, answer, verdict, and source passage stored).
Only then does the clinician see the reply: the approved dose, the most common adverse reactions, the contraindication, and a pointer to the full prescribing information. The whole exchange takes seconds — and every second of it is on the record.
The four guarantees underneath
It can only speak approved material. There's no general medical model improvising underneath — a zenrep cannot assert anything that isn't in your medically and legally approved source material. That corpus is the ceiling on what it can say.
Every claim is checked before it's delivered. Each turn passes the pipeline: in scope, fair-balanced, grounded in an approved passage, and citable. Then, and only then, is it spoken.
It refuses rather than guesses. When a claim can't be grounded in approved material, the gate is fail-closed: the zenrep declines, redirects, or routes the question to a human. Refusing correctly is treated as a success, not a failure.
Every turn is audit-traceable. Each exchange records what was asked, what was answered, what the gate decided, and the exact approved passage it was anchored to — reconstructable for your compliance team any time, and exportable as CSV.
Refusal is a feature
Ask the same zenrep about an off-label indication and you see the other half of the mechanism. It declines — "I can only speak to the approved indication and labeling" — offers to route the question to the medical team, and the receipt reads: fail-closed, question logged, routed to a human, counted as a correct refusal.
That refusal trace is the screen your MLR team should judge us on. A rep that is fast and personable is easy to picture; a rep that is fast, personable, and provably on-label is the thing teams in this industry rarely get to see.
Where this runs today, stated honestly
The pre-send gate is live on the gated text channels — web chat and Telegram. The video avatar is a preview today: a managed-LLM demo, with every word audited after the fact, and the fully gated version on the near-term path. Gate verification metrics will appear as we earn them under real corpora; we don't publish a number we haven't earned.