Workflow Risk, powered by AI Syndicate.

Assess AI workflow risk before it acts.

Before you let AI workflows take action, score the risk. Answer 12 questions and get a risk tier, guardrail recommendations, and a one-page report.

Run Free Risk Check See How to Enforce Controls

Sample output

Tier 3 - Controlled Action

Why this tier was assigned

The workflow can update records within a narrow scope. It has partial approval controls and should stay bounded until enforcement is in place. This workflow needs pre-execution enforcement before broader autonomy.

Pre-execution approval gating

Identity binding

Policy binding

Runtime evidence

Budget and rate limits

Fail-closed enforcement

Higher-risk actions need control before execution.

See how AI Syndicate enforces this at runtime

4-tier model

Not all AI actions are equal.

Higher-risk actions need control before execution. Workflow Risk classifies AI workflows by what they can do, how much autonomy they have, and which controls are already in place.

See how AI Syndicate enforces this at runtime

Tier 1

Educational

Informational

AI reads, summarizes, or suggests only. No external action capability.

Tier 2

Educational

Assisted

AI drafts or prepares actions, and a human approves before every action.

Tier 3

Product

Controlled Action

AI can take bounded actions with low or medium impact and partial controls.

Runtime controls should enforce approvals, policy, and evidence before execution.

Tier 4

Product

High-Risk Autonomous

AI can take high-impact, irreversible, or broadly scoped actions with weak controls.

Runtime controls are required before production autonomy.

How it works

Fast enough for intake. Clear enough for review.

Start assessment

Step 1

Answer 12 questions

Capture the workflow, actions, autonomy, and controls in a short assessment.

Step 2

Get a risk tier

A deterministic rules engine assigns one of four action-risk tiers.

Step 3

Review required controls

See the controls needed before the workflow can safely act.

Step 4

Share the report

Open a printable report for security, platform, or leadership review.

Built for teams deploying AI that can take action.

Designed for security, platform, engineering, and innovation leaders who need a defensible control path before workflow autonomy expands.