Detect Deepfakes automatically
Our AI deepfake detector automatically detects image and video manipulations such as face swaps.
"status": "success",
"request": {
"id": "req_fPDJ8fsQGgC1E2Ed8V",
"timestamp": 1709634469.399,
"operations": 1
},
"type": {
"deepfake": 99%
},
"media": {
"id": "med_fPDJiQzaZR2E193CXb",
"uri": "image.jpg"
}
Detect Deepfakes automatically
Likely AI-generated
Likely Deepfake
Not likely to be AI-generated or Deepfake
Uncertain if AI-generated or not
Uncertain if Deepfake or not
Detect deepfakes to prevent manipulation, fraud, abuse and spam.
Mis- & Disinformation
Spreading deepfakes in media to manipulate public opinion.
Fake profiles
Scamming and extorting users with fake profiles on social networks.
Fake IDs
Forging ID documents to bypass onboarding checks or take over accounts.
Impersonation
Misleading viewers by representing someone without their consent.
Nudification
Creating fake nude images of people without their consent.
Fake news
Spreading fake news to harm democratic processes.
Specialized models to detect Face Manipulations.
Our AI models are trained to analyze the outline and consistency of faces in your media.
Our models use advanced face analysis to detect alterations, including face swaps, face regenerations and AI-generated modifications, ensuring authentic facial representations.
The models can be easily integrated into your existing processes: during user onboarding, content moderation, or media analysis. Processing is done in near real-time, thereby reducing friction and ensuring a seamless user experience.
Powerful models to detect AI-generated media.
Our Automated AI-generated media detection models can help you determine the provenance and authenticity of images.
Our models are trained to detect images generated by popular Diffusion models such as Stable Diffusion, Firefly, Dall-E, Ideogram..., GANs and other AI generators.
Those models can be used in conjunction with the face manipulation detection.
What is a Deepfake?
Deepfakes come in many shapes and forms, and can serve different purposes. Most deepfakes are generated using a combination of the following 3 types of actions:
AI-generation
AI-generated images and videos are created by AI models such as Diffusion models (Stable Diffusion, MidJourney, Flux...) or GANs.
They are generated out of the blue, or can take a real image as input and modify it.
AI-models can also be used to modify parts of an image (inpainting) or extend it (outpainting).
Face swaps
Face swaps are a common form of deepfakes. They consist of taking a real image or video of a person, and replacing their face with another person's face.
AI models are used to adapt the angle, lighting and expressions of the face to match the provided image or video. Results can be extremely convincing.
Face manipulations
Face manipulations are the result of AI algorithms modifying a person's face, either to change its likeness or to change its emotions and lip movements.
This is what so-called lip-sync models use to adapt a person's face to a made-up or modified speech.
Part of our Integrated Content Analysis Suite.
We offer a complete set of models and tools to analyze and moderate your visual media: Content Moderation, Content Analysis, Media retrieval.
Image Moderation
Moderate user-generated images and pictures. Detect any unwanted content such as adult content, hate, violence, children, and more. More on our Image Moderation API
JPEG PNG WebP GIF...
Video Moderation
Analyze static videos and live-streams in near realtime. Flag unwanted content such as violence, adult content, illegal imagery and media that violate your terms of use. More on our Video Moderation API
MP4 AVI MOV HLS RTSP...
See why the world's best companies are using Sightengine
Empower your business with Powerful Content Analysis and Filtering.
GET STARTEDSEE DEMO