Multi-modal AI content detection covering text, images, video, and audio. Trained to catch humanized AI content and deepfakes, with sentence-level confidence scores.