Skip to main content

AI-Generated Media Recognition

J
Written by Jules Plantec

AI-Generated Media Recognition helps you identify whether submitted images or videos may have been created using generative AI.

SnapCall analyzes the media and returns detection scores across a wide range of image and video generation models, giving your team an additional signal to support fraud prevention, authenticity checks, and automated decision-making.
​

How to activate AI-Generated Media Recognition

AI-Generated Media Recognition can be activated directly from your Flow configuration.

To enable it:

  1. Open the Flow Configurator.

  2. Go to the General tab.

  3. Find AI-Generated Media Recognition.

  4. Enable the feature for your Flow.
    ​
    ​

Once activated, SnapCall will automatically analyze submitted media for signals associated with supported AI-generation models.

[Insert screenshot here: Flow Configurator → General tab → AI-Generated Media Recognition]

AI-Generated Media Recognition can be enabled from the General tab of the Flow Configurator.
​

What does AI-Generated Media Recognition do?

For each submitted media, SnapCall looks for signals associated with known AI-generation models.

You can use the result to:

  • identify potentially AI-generated content

  • detect the most likely generator

  • flag suspicious submissions for review

  • trigger specific workflows or business rules

  • combine AI-generation detection with other AI Skills

  • use the result in custom fields

What does the result look like?

SnapCall returns a percentage score for each supported generator.

For example:

  • midjourney: 82%

  • dalle: 6%

  • adobefirefly: 0%

  • other_image_generators: 12%

A higher percentage indicates a stronger match with signals associated with that generator.

This makes it possible to quickly identify which AI model is most likely associated with the submitted media and use that information in your workflow.
​

Which generators are supported?

SnapCall can analyze signals associated with a broad range of image and video generation technologies, including:

  • stablediffusionxl

  • leonardo

  • stablediffusion

  • flux

  • sdxlinpaint

  • adobefirefly

  • dalle

  • midjourney

  • ideogram

  • mai

  • bagel

  • ray3

  • omnigen

  • imagen

  • meta

  • other_image_generators

  • kling

  • 4o

  • grok

  • wuerstchen

  • qwen

  • wan

  • pika

  • reve

  • ltx

  • bria

  • krea

  • lucid

  • stablediffusioninpaint

  • kandinsky

  • dmd2

  • longcat

  • grokimagine

  • flux2

  • ovis

  • titan

  • higgsfield

  • cosmos

  • runway

  • moonvalley

  • var

  • imagen4

  • pixart

  • gan

  • recraft

  • infinity

  • zimage

  • bingimagecreator

  • gptimage1_5

  • hidream

  • lcm

  • mcnet

  • glide

  • sora2

  • veo3

  • blip3o

  • stablecascade

  • sana

  • mochi

  • hunyuan

  • cogview

  • liveportrait

  • sanavideo

  • transpixar

  • pyramidflows

  • sadtalker

  • janus

  • gemini3

  • luminagpt

  • deepfloyd

  • vibe

  • hailuo

  • luma

  • steadydancer

  • seedream

  • vqdiffusion

  • sora

  • aniportrait

  • emu3

  • gemini

  • hallo

  • viduq2

  • amused

  • cogvideos

  • makeittalk

  • switti

  • imagineart

  • heygen

  • hedra

  • seedance

  • personalive

  • pixverse

  • dreamid

  • flashvideo

  • scail

  • haiper

What if no generator is detected?

If all generators return 0%, SnapCall did not identify signals associated with any of the supported AI-generation models.

This provides your team with an additional confidence signal when reviewing submitted media. The result can also be combined with other AI Skills, business rules, or Resolution to support your decision-making.
​

How can I use AI-Generated Media Recognition in my workflow?

The detection result can be used as part of your wider SnapCall workflow.

For example, you can:

  • automatically flag media with a high AI-generated score

  • route suspicious submissions for manual review

  • store the detected generator and score in custom fields

  • combine the result with damage, document, or eligibility analysis

  • use Resolution to recommend the appropriate next step

This allows AI-generated media detection to become part of a broader automated review process rather than a standalone check.

Related Articles
Did this answer your question?