Deployment Safety Policy

Syntic's commitments around the safe deployment of an AI Workforce. This policy describes how we evaluate AI Employee capability, what escalation tiers we operate, and how the platform stays trustworthy as autonomy scales.

Capability monitoring

Every AI Employee and AI Agent shipped through Syntic is continuously evaluated against our internal capability bar. Significant jumps in capability trigger pre-deploy review before rollout reaches production Workforces.

Escalation tiers

Syntic operates four deployment tiers — Sandbox, Pilot, General Availability, and High-Autonomy. Movement between tiers requires explicit safety review, expanded evaluation coverage, and customer-facing disclosure of what changed.

Third-party audits

Independent safety auditors review the Syntic Workforce on a published cadence. Audit summaries are published in the quarterly Transparency Report; full reports are available under NDA.

Customer commitments

Customers retain control over which tier each AI Employee runs at, what data the AI Workforce sees, and which Skills it can invoke. Syntic does not silently expand AI Employee scope without customer-side opt-in.