I used to be annoyed by the huge influx of hoaxes, fake news, and especially fake quotations long before any LLM was born. And today’s chatbots can hallucinate quotations for you, among other false information. And deepfakes showed up before the chatbot mania. However, social networks are invaded today by AI-generated images, sometimes with no harmful purpose but other times in order to create an audience for false accounts.

On the harmless side, I’m almost always able to detect a fake image. Today, I notice the success of an image meant to promote the concept of “formaggeria-salumeria italiana.” It gained momentum in the last two months across social networks. People were salivating and craving such products, wherever they were in the world.

Except that, just as I thought, it was a fake. Regardless of the size, the text was unclear and more like a mix of a Georgian script with something else.

Sightengine’s AI image detector can’t be fooled. And their free tier is quite generous.

This is a business field that should become increasingly successful in time, in a world infested with manipulation, deepfakes, fraud, abuse, spam… Where Sightengine helps:

  • Mis- & Disinformation: Spreading AI-generated mis-information and deepfakes in media.
  • Fake insurance claims: Creating fake insurance claims with faked images of incidents at home.
  • Fake profiles: Scamming users with fake profiles on social networks.
  • Fake IDs: By-passing KYC/AML checks with fake or spoofed identity documents and selfies.
  • Marketplace spam: Spamming and flooding marketplaces with auto-generated variations of images.
  • Fake photo evidence: Generating fraudulent accident reports with AI-generated images.
  • Impersonation: Misleading viewers by representing someone without their consent.
  • Nudification: Generating abusive images to harm users through impersonation or nudification.
  • Fake news: Spreading fake news to harm democratic processes.

They can help identify images across AI image generators such as: MidJourney, DALL-E, Stable Diffusion, Ideogram, Flux, Bing Image Creator, GANs.

BTW, this is a French business, with correspondence to go to boîte 37, 16 bis Rue d’Odessa, 75014 Paris, France. There is a multi-storey building at the address, a cinema (Pathé Montparnos), a Lebanese restaurant, and more.

For other fields, there are many other tools to detect AI-powered fraud, especially plagiarism.

Welcome to the brave new world of AI!