Pre-action verification by Inspect IQ

Verify IQ

Trust, but verify.

Before you act, verify what matters.

Verify IQ is a bounded pre-action verification capability for suspicious or consequential requests. It is designed for the moment before a user takes an action that may be difficult to reverse.

Decision boundary

What can the available evidence support — and what remains unresolved?

Use moments

Pause at the consequential moment

A supplier says payment details changed.

A message directs you to sign in through a destination or link.

Someone claiming to be support asks for an authentication code.

You want to interpret supplied media provenance or Content Credentials information.

Public tool scope

Three bounded verification tools

01

analyze_suspicious_content

Examine supplied suspicious content and identify supported warning signals, uncertainties and safer next steps.

02

verify_entity_or_destination

Assess supplied details about a claimed entity or destination before a user pays, signs in or shares sensitive information.

03

inspect_media_provenance

Interpret caller-supplied media provenance and Content Credentials information without claiming independent authentication of the media bytes.

Clear limits, useful answers

Verify IQ describes what the supplied evidence can support without converting uncertainty into false confidence.

  • No universal fraud or scam score.
  • No safety or legitimacy guarantee.
  • External and live verification sources are disabled in the current review build.
  • Media provenance interpretation uses caller-supplied provenance information; it does not independently authenticate arbitrary media bytes.
  • Missing Content Credentials does not prove manipulation.

Current status

Functional beta/testing review candidate for Meta AI Connectors developer preview.

This status does not indicate Meta approval, partnership, endorsement or production readiness.