Free PSD colorful gear vr mockup via Freepik
AI-EQUIPPED EYEGLASSES READ LIPS!
Cornell University researcher Ruidong Zhang from the Smart Computer Interface for Future Interactions (SciFi) Lab is the author of “EchoSpeech: Continuous Silent Speech Recognition on Minimally Obtrusive Eyewear Powered by Acoustic Sensing.” For people who cannot vocalize sound, this silent speech technology could be an excellent input for a voice synthesizer. It allows communication through a smartphone and helps synthesize voice from AI-powered acoustic-sensing based on lip movement. “It could give patients their voices back,” Zhang said for the future of EchoSpeech. (bit.ly/AIGlasses)
A BIO-INSPIRED SENSOR DETECTS MOVING OBJECTS FROM VIDEO
Aalto University has found an alternative to the present motion detection that is component heavy and needs analysis to be done one frame at a time. This neuromorphic vision technology can integrate several optical images in one frame and remember movement with AI neural networks to detect motion and predict its path. Following many tests, it can use this technology for intelligent, self-driving transport. Research fellow Hongwei Tan said, “The result is a compact and efficient sensing unit.” (bit.ly/BioSensors)
GPT-5 TRAINING HALTED
At MIT, Sam Altman announced OpenAI will pause the fifth-gen GPT training for six months and focus on the GPT-4 model. He advised the developers to pause the fifth-gen GPT as OpenAI seeks to address concerns and safeguard its AI model’s safety and ethical ramifications. “We are doing things on top of GPT-4 that I think have many safety issues that we need to address,” said Sam Altman. (bit.ly/GPT 5 Training)
GOOGLE ATTEMPTS TO COUNTER BING
Microsoft integrated OpenAI’s GPT into its search engine and set Google off on a drive to create enhanced AI-powered features for its search engine under a project coined ‘Magi.’ They slotted a public release for this month with limited capabilities that will offer a more ‘personalized experience.’ It is possible Samsung is thinking of switching to Bing as its default search engine, but Lara Lavin from Google said, “We’re excited about bringing new AI-powered features to search, and will share more details soon,”(bit.ly/Google AI)
COPYRIGHT CASE MIGHT TRANSFORM GENERATIVE AI
The US Copyright Office recently stated that it cannot offer protection for AI-generated art under copyright laws. The US Copyright Office allows only art that contains a ‘significant human authorship’ to be copyrighted. With lawsuits filed against Midjourney, DeviantArt, and Stability AI for infringement, they are accused of training generative models on the work of countless artists. Everyone is asking the fundamental question about ‘Fair Use’ and if generative models trained on copyrighted works can be covered. When does the act of creation really begin? (bit.ly/Generative AI) 
