Confidence & escalation

Confidence thresholds determine when DelegateZero is allowed to act automatically. Each decision is assigned a confidence score between 0 and 1. If the score meets or exceeds your configured threshold, the decision may be executed. Otherwise, it is escalated or returned as a draft. Lower thresholds increase autonomy. Higher thresholds increase safety.

You can override confidence thresholds on a per-request basis using instructions.confidence_threshold.

Policies & safety

Policies define non-negotiable rules and boundaries that DelegateZero must follow. They always take precedence over confidence, precedents, and playbooks.

If a request conflicts with a policy, DelegateZero will escalate instead of guessing or partially complying.

Use policies to encode rules such as approval limits, forbidden actions, compliance requirements, or escalation conditions.

Hard stops are conditions under which DZ must never act automatically. When a hard stop is triggered, the request is escalated regardless of confidence.

Examples include irreversible actions, legal or financial risk, or missing required context, and these can be configured through policies.

Risk levels

Risk levels allow you to categorize decisions by their potential impact. Higher-risk categories can require higher confidence thresholds, stricter policies, or mandatory escalation.

Risk levels are useful for gradually increasing autonomy as trust is established.

Assistant attribution

Assistant attribution controls how responses are framed when DZ communicates externally. You can choose whether responses appear to come directly from you or from an assistant acting on your behalf - a reasonable next-step on your path to feeling confident with DZ being able to handle your decision-making.

This setting affects tone and attribution only. It does not change the underlying decision logic.

You can override attribution for individual requests using instructions.attribution.

Confidence Autopsy

Confidence Autopsy is a weekly digest that surfaces patterns across your escalations. Rather than reviewing individual escalations one at a time, Confidence Autopsy identifies the underlying causes - missing context, policy conflicts, recurring gaps - and shows you the single change most likely to resolve the most escalations.

You receive this as a weekly email. It is also accessible in the dashboard under the Autopsy tab.

Use Confidence Autopsy to prioritize context improvements. It is the fastest way to reduce escalation rate and increase the quality of autonomous decisions over time.

Confidence Autopsy is available on the Operator plan and above.

Decision Simulation

Decision Simulation lets you test context changes before committing them. Before adding a new policy, updating a playbook, or changing any context entry, you can run a batch of your historical decisions against the proposed change and see how outcomes would have differed.

To run a simulation: navigate to Settings > Simulation, select the context change you want to test, choose a date range of historical decisions to replay, and run. Results show a side-by-side diff of original vs. simulated outcomes.

This is especially useful when: rolling out a new policy, onboarding a team member to a decision domain, testing compliance changes, or evaluating the impact of a context update before it affects live decisions.

Decision Simulation is available on the Operator plan and above.

Judgment Profiles

A Judgment Profile is a compiled, portable snapshot of how a specific person makes decisions. It is built automatically from your Memory, Policies, and Precedents over time and can be exported as a versioned artifact.

Use cases:

• Onboarding a new decision-maker - export your Judgment Profile and import it into a new team member's workspace to seed their context with your established judgment before they take over a decision domain
• Spinning up a new workspace - import your profile into a new DelegateZero account to carry your decision patterns forward without rebuilding from scratch
• Delegation - share a domain-specific profile with a colleague so their DelegateZero configuration reflects the judgment they've inherited

A Judgment Profile is a point-in-time snapshot. It does not sync automatically after export. The recipient's Memory and decisions after import will diverge from yours over time, as it should.

To export a profile: navigate to Settings > Judgment Profile > Export. To import: Settings > Judgment Profile > Import.

Judgment Profiles are available on the Team plan and above. Importing a profile requires a Team plan or above.

Delegation Chains

Delegation Chains allow decision authority to flow across multiple users in a hierarchy, with DelegateZero holding the logic at each handoff.

A user at the top of the chain sets policies that apply to everyone below. Each user below can extend the chain with their own playbooks and precedents, which layer on top of upstream policies without overriding them. When a user in the chain overrides a decision, the override is scoped to their layer only - upstream policies remain intact.

This makes DelegateZero an organization-level system rather than an individual tool. Each user in a chain operates with their own configured layer, and DelegateZero enforces the full stack at decision time.

To configure a chain: navigate to Settings > Delegation and add chain members. Each member must have a DelegateZero account.

Delegation Chains are available on the Team plan and above.