Panasonic HD's OmniFlow AI enables seamless conversion between text, images, and audio, reducing data costs and improving efficiency.
Panasonic Holdings Co., Ltd. (Panasonic HD) and Panasonic R&D Company of America (PRDCA), in collaboration with UCLA researchers, have developed OmniFlow, a multimodal generative AI capable of converting between text, images, and audio ("Any-to-Any").
OmniFlow reduces training data costs by combining specialized generative AI models (text-to-audio, text-to-image) and learning high-precision models with minimal multimodal data. This breakthrough was presented at CVPR 2025.
AI companies pay hundreds of thousands for failed startups' workplace data from Slack, Teams & Jira to train...
Wikipedia banned AI language models like ChatGPT in March 2026 to combat misinformation and Russian disinformation...
Wikipedia's human-curated knowledge has become essential for AI systems to avoid model collapse. The platform's...
WeTransfer removed AI training clauses from its terms after media reports, claiming they never used files for AI....
CSC Surf and Nokia reach a 12 Tbit/s data transfer milestone, paving the way for advanced AI supercomputing.
March 2026 Iranian drone strikes on AWS data centers in UAE and Bahrain mark the first kinetic attack on hyperscale...
Follow Discussion