---
sourceDocument: Brazil Enable AI
sourceDocumentLink: https://www.servicenow.com/docs/r/intelligent-experiences

 Release :

    - brazil

ft:locale :

    - en-US

ft:publication_title :

    - Brazil Enable AI

ft:clusterId :

    - platai

bundleId :

    - platai

workflow :

    - Platform


---

# Reviewing prediction errors with the Observability Dashboard

# Reviewing prediction errors with the Observability Dashboard {#ariaid-title1}

Release version: Brazil  
Updated September 10, 2026  
![](https://www.servicenow.com/docs/portal-asset/ico-clock) 1 minute to read  
The Observability Dashboard offers a unified view and actionable insights for errors detected in Predictive Intelligence. Use this dashboard to visualize logged errors and gain information on prediction reliability and potential problem areas.  
View the PI - Observability Dashboard by navigating to Predictive IntelligenceDashboardsObservability. The dashboard contains the following widgets.

* Total Number of Prediction Errors
* Prediction Errors Breakdown by Date
* Prediction Errors Count by Capability
* Prediction Error Count by Error Type
* Error Types by Capability
* Successful and Unsuccessful Predictions Breakdown by Date

{#prediction-errors-observability-dashboard__ul_fp4_nzl_1gc}You can drill down to the underlying records from the widget graphics. You can also change the date range of all widgets by selecting Date to open the selector.  
Figure 1. PI Observability Dashboard --- upper four widgets  
The PI - Observability Dashboard draws from a table dedicated to logging prediction errors: ML Predictor Error Logs \[ml_predictor_error_logs\].

* Fields in the table include Error Type, Error Message, Status code, Solution, and Capability.
* View this table's records directly by entering <kbd class="ph userinput">ml_predictor_error_logs.list</kbd> in the navigator.
* Roles required to access the table: ml_report_user or ml_admin.
{#prediction-errors-observability-dashboard__ul_bdv_xxl_1gc}  
The table logs the following types of granular errors.

* Client-side issues (400 Series) --- captures request errors such as timeouts and invalid inputs.
* Server-side issues (500 Series) --- tracks internal server errors encountered during prediction.
* Internal prediction failures --- identifies instances when the model was unable to generate a prediction.
* Low confidence predictions --- records log results falling below a defined confidence threshold.

{#prediction-errors-observability-dashboard__ul_blr_rzg_bgc}  
Note:  
Errors in training aren't included in this table. For a dashboard reporting on training errors, see [Predictive Intelligence Usage Analytics dashboard](https://www.servicenow.com/docs/oeyDThWBjAPv4J2GPS1sHQ "The Predictive Intelligence Usage Analytics dashboard is a central location to understand the effectiveness and overall value of all your Predictive Intelligence solutions. View metrics for model training successes and failures. Monitor prediction statistics including breakdowns by individual model.").  
Figure 2. PI Observability Dashboard --- lower two widgets

*[\>]: and then