deepfakes
Liquid AI's Real-Time Audio Model Enables Live Deepfakes
LFM2-Audio-1.5B achieves sub-100ms response latency for speech synthesis, opening doors for real-time voice cloning and live audio deepfake applications.
Latest
ai-infrastructure
OpenAI Partners with Samsung, SK Hynix for Stargate
OpenAI secures deals with Samsung and SK Hynix for DRAM supply and South Korean data centers, advancing the Stargate AI infrastructure project.
ai-video
Samsung Powers OpenAI's Stargate with 900K Chips Monthly
OpenAI partners with Samsung and SK Hynix to secure 900,000 semiconductor wafers monthly for Stargate project, enabling next-generation AI video and synthetic media capabilities.
deepfakes
AI Voice Clones Cross the Uncanny Valley
New research reveals AI-generated voices have become indistinguishable from human speech, marking a watershed moment for synthetic audio and authentication challenges.
ai-video
OpenAI Launches Sora 2: The ChatGPT Moment for Video
OpenAI unveils Sora 2, its new AI video and audio generation system, creating deepfakes so realistic that distinguishing them from reality becomes nearly impossible.
ai-video
How AI Models Generate Videos: Technical Deep Dive
MIT Technology Review explores the mechanics behind AI video generation while uncovering bias issues in OpenAI's Sora model affecting synthetic media authenticity.
deepfakes
Gartner: 62% of Firms Face Deepfake Attack Crisis
New Gartner research reveals 62% of organizations have experienced deepfake attacks, highlighting the urgent need for detection and authentication technologies.
deepfakes
Darknet Deepfake Services Drop to Alarming Price Points
Kaspersky research reveals deepfake creation services on darknet markets now cost as little as $10, making sophisticated synthetic media attacks accessible to anyone.
deepfakes
AI Voice Clones Cross Uncanny Valley Threshold
New research reveals AI-generated voices have reached a critical milestone where listeners can no longer reliably distinguish synthetic speech from human recordings.
ai-video
AI Stan Lee Hologram Sparks Digital Ethics Debate
LA Comic Con's $20 AI-powered Stan Lee hologram draws criticism as fans question the ethics of synthetic avatars representing deceased celebrities.
ai-security
CAMIA Attack Exposes AI Memory Vulnerabilities
New privacy attack method CAMIA reveals whether your data was used to train AI models, exposing critical vulnerabilities in synthetic media generation systems.
ai-video
New Physics-Aware Training Makes AI Videos More Realistic
PIRF technique improves diffusion models by enforcing physical laws during generation, potentially creating more convincing synthetic videos that follow real-world physics.