ServiceNow AI Lens limitations

  • Rversion finale: Australia
  • Mis à jour 12 mars 2026
  • 1 minute de lecture

    • Be aware of a few limitations when you’re using the ServiceNow AI Lens application.

    Tableau 1. Limitations with ServiceNow AI Lens
    Limit Description
    Maximum number of images Maximum number of images that can be uploaded to ServiceNow AI Lens is 10.
    Maximum character limit In the instructions field for the analysis, you can enter up to 500 characters.
    Themes When you change the theme on the ServiceNow instance, the theme of the ServiceNow AI Lens is automatically updated. But, if you have already started a ServiceNow AI Lens session and then you change the theme on the instance, the theme doesn't change automatically until you launch a new session.

    This behavior is standard for the ServiceNow AI Platform.
    Language When you change the language on the ServiceNow instance, the language of the ServiceNow AI Lens is automatically changed. But, if you have already started a ServiceNow AI Lens session and then you change the language on the instance, the language doesn't change automatically until you launch a new session.

    This behavior is standard for the ServiceNow AI Platform.

    When using ServiceNow AI Lens to scan non-English documents—such as invoices in Chinese— you may notice inconsistencies in translation, formatting, or data extraction. These discrepancies result from limitations in how the underlying LLM processes different locales.
    Auto-fill Pre-populated form fields, such as record number, can be overwritten by ServiceNow AI Lens. You must verify the auto-filled form fields before saving the form.
    Visibility of the Create with Lens button As a user with the lens_user role, if you don’t have the Write access for the first record of any list, the Create with Lens button doesn't appear in the list view of the table. You can sort the records to show a different record on top and reload the list to view the Create with Lens button.
    Hallucinations If the large language model (LLM) returns text that doesn’t exist in the source artifacts, the LLM might inject out-of-context information. You must verify that all information aligns with the source material to ensure that there are no fabricated facts or unsupported conclusions.