AI Detection
New research introduces PADBen, a comprehensive benchmark evaluating how AI text detectors perform against paraphrase attacks. The framework reveals critical vulnerabilities in current detection systems through adversarial testing.
deepfakes
Medical professional becomes latest victim of deepfake technology, falsely endorsing weight loss products in AI-generated video. Case highlights growing threat of synthetic media in fraudulent health product marketing.
deepfakes
A manipulated video falsely claiming President Droupadi Murmu was blackmailed has been identified as a deepfake, highlighting the growing challenge of synthetic media targeting political figures in India.
LLM
Large language models sometimes generate plausible-sounding but false information. Understanding the technical causes of AI hallucinations is crucial for building reliable synthetic media systems and detecting AI-generated misinformation.
deepfakes
AI-generated deepfake videos impersonating prominent physicists spread false claims about an interstellar object, demonstrating sophisticated misuse of synthetic media to amplify scientific misinformation.
Multimodal AI
Large language models struggle with long documents due to context window limits. New research shows converting text to images before processing could dramatically improve AI's ability to handle vast amounts of information.
Generative AI
Researchers introduce mathematical framework for understanding how generative AI models create intermediate outputs between training data points, with implications for video generation and synthetic media quality.
AI music
Universal Music Group settles copyright lawsuit with AI music generator Udio, forming partnership instead. The case centered on training data usage and represents a significant shift in music industry's approach to generative AI.
voice cloning
Researchers demonstrate AI can clone voices using just photographs, eliminating the need for audio samples. This breakthrough raises new concerns for synthetic media and digital authenticity verification.
AI Video
YouTube is giving users control over AI upscaling that automatically enhances low-resolution videos. The new toggle lets viewers disable the enhancement technology when they prefer original quality over AI-processed content.
Adobe
Adobe announces expanded AI capabilities across Photoshop, Premiere Pro, and Lightroom at MAX 2025, including Generative Extend for video footage and AI-powered text generation that works across multiple image layers simultaneously.
deepfakes
New research reveals how generative AI is being weaponized to create synthetic vehicle damage images, enabling a surge in fraudulent insurance claims.