Glossary
Trust Threshold
A trust threshold is the minimum Signet Score or dimension score that an AI agent must meet to be eligible for a specific transaction type, platform access level, or operational privilege.
What is Trust Threshold?
Not all interactions require the same level of trust. Browsing a product catalog requires minimal trust. Processing a $10 payment requires moderate trust. Executing a $1 million investment requires very high trust. Trust thresholds formalize these risk-proportional requirements by setting minimum score requirements for different activities.
Thresholds can be set at multiple levels. Platform-wide thresholds might require a minimum composite score for any agent to operate on the platform. Transaction-level thresholds might require higher scores for higher-value transactions. Dimension-specific thresholds might require a minimum Security score for agents handling sensitive data, regardless of their composite score. This multi-level approach enables fine-grained access control that matches trust requirements to actual risk.
Well-designed trust thresholds balance security with accessibility. Setting thresholds too high excludes legitimate agents and limits platform utility. Setting them too low admits risky agents and undermines platform trust. The optimal thresholds depend on the specific risk profile of each activity and should be calibrated based on observed outcomes -- adjusting upward if losses are too high and downward if qualified agents are being unnecessarily excluded.
Example
An enterprise procurement platform sets tiered trust thresholds: agents need a composite score of 500+ for browsing vendor catalogs, 650+ for submitting quotes, 750+ for executing purchases under $10,000, and 850+ with a Financial dimension minimum of 800 for purchases over $10,000. These thresholds ensure that the trust bar rises with the stakes of the activity.
How Signet addresses this
Signet's recommendation tiers (Clear 700+, Review 400-699, Caution below 400) provide default trust thresholds that platforms can adopt directly. For more nuanced needs, platforms can set custom thresholds using Signet's API, filtering by composite score, individual dimension scores, confidence tiers, or any combination. Signet provides analytics on threshold effectiveness, helping platforms calibrate their thresholds for optimal security-accessibility balance.
Build trust into your agents
Register your agents with Signet to receive a permanent identity and trust score.