Independent image-authenticity product
What Made It? helps people inspect what an image file can—and cannot—support. The product keeps visual patterns, file evidence and available provenance separate so a user can challenge the result instead of trusting an unexplained percentage.
What Made It? accepts JPEG, PNG and WebP images. It validates the file, reads available metadata and structure, checks available Content Credentials and prepares an evidence report. The controlled detector Beta can combine independent self-hosted and provider channels; it labels one-channel results as lower confidence and withholds a direction when the usable evidence is missing or conflicts.
The What Made It? product and methodology team maintains the product interface, Methodology, Research and release-status language. Updates are versioned against the deployed processing configuration so a public page does not claim that a detector or provider is active when it is disabled.
An assessment does not prove authorship, intent, copyright ownership or legal admissibility. Missing metadata or missing Content Credentials is not proof of AI generation. The product should not be the sole basis for employment, legal, financial, moderation or safety decisions.
Accountability
Methodology explains evidence and decision limits. Research and release status states what is currently enabled. Privacy describes processing and retention. Terms describes permitted use and billing conditions.
Product support, privacy requests and security reports can be sent to support@whatmadeit.com. Transactional account email is sent separately from accounts@whatmadeit.com.
About and accountability information last updated