Deepfake video injection in eKYC: Scam AI and ALCHERA
Deepfake video injection in eKYC: ALCHERA's face checks plus a Scam AI score on each frame of the selfie video, so an injected deepfake is held for review.
Read storyCollaboration in practice
Explore how we work together, starting with the problem you want to solve.
Meet the companiesDeepfake video injection in eKYC: ALCHERA's face checks plus a Scam AI score on each frame of the selfie video, so an injected deepfake is held for review.
Read storyVoice deepfake detection for AI voice agents on KrosAI lines: in our demo, a synthetic caller asking for a PIN reset was flagged and a real voice passed.
Read storyAI content in consumer insights: Honestly labels the opinions and Scam AI can score their pictures, so a generated image is weighed, not deleted.
Read storyVoice deepfake detection for payments: in a Gallup survey, 70% of scams with a phone call also used texts. Modulate checks the call, Scam AI the invoice.
Read storyLayered deepfake detection for KYC: in deepidv's benchmark, 31% of GAN faces got past a single-layer liveness check and 0.2% past its multi-layer stack.
Read storyWhitney AI post-trained a model on Scam AI's real scam calls to write synthetic scam call data; model judges took its turns for real 41.3% of the time.
Read storyShort-term rental booking fraud returns weeks later as a chargeback and a $15 Stripe fee. Scam AI and Cozystay score each booking and check in risky guests.
Read storyAn AI scam call honeypot whose personas speak in Fish Audio voices, and what its first 10,211 recorded calls show fraud teams about scam scripts and requests.
Read storyOn-device deepfake detection for video calls: Halo checks faces on the Snapdragon NPU and uploads nothing. A 2024 deepfake call cost Arup about $25.6 million.
Read storyDeepfake candidate detection in video interviews: Halo checks an Eteon candidate's face live, and our API checks the voice in short recorded answers.
Read storyScam AI × Aident: an AI agent checks the photos, video and recordings in a claim for AI generation before relying on them, and a person reviews what is flagged.
Read storyHow the Scam AI and TinyFish integration checks hiring candidates and insurance claims: public pages bound to the case, files checked, gaps written down.
Read storyWork with Scam AI
Have a collaboration in mind?