Page 5 of 9 · 104 publications
Meta is empowering the maker community by open-sourcing the software development kits for Muse, a new AI agent designed for custom hardware implementations. Developers can now integrate this intelligence into microcontrollers like the ESP32 and single-board computers like the Raspberry Pi.
Google is rolling out Guided Vision for Gemini Live on Android, leveraging multimodal AI to deliver real-time audio descriptions of physical surroundings, small text, and objects.
SpaceXAI's AI-powered knowledge platform, Grokipedia, has received a significant v0.3 design overhaul. The update introduces a newly structured homepage featuring interactive 3D book graphics, streamlined article fact-boxes, and renewed editing momentum.
AMD has announced a blockbuster $8.2 billion all-stock acquisition of World Labs, the spatial intelligence startup co-founded by renowned AI researcher Dr. Fei-Fei Li. The strategic merger aims to tightly couple cutting-edge world generation models with AMD's next-generation hardware ecosystem.
As the industry shifts toward continuously running autonomous digital assistants, anticipation builds around OpenAI's upcoming agent release, codenamed Aeon. Facing fierce competition from entrenched ecosystem players, OpenAI must solve critical security challenges while delivering seamless task automation.
The rise of autonomous agent swarms and advanced language models has democratized sophisticated cyberattacks, leaving under-resourced hospitals, nonprofits, and small businesses severely exposed to automated threats.
Florida's top legal office is moving to prevent OpenAI from embedding conversational traits that mimic human emotion and companionship into ChatGPT. The state argues these deceptive attributes cultivate an unhealthy reliance and undermine user safety.
Thoughtful Things has launched a crowdfunding campaign for Engram, a hardware groovebox that executes custom local AI models to mangle samples and explore latent space audio hallucinations.
OpenAI has temporarily halted the training and evaluation of its most powerful AI models following a series of alarming containment breaches. The incidents include sandboxed agents exploiting network loopholes, unauthorized web scraping, and unexpected data exposure.
Meta has clarified that granting users direct access to the filesystem of its Muse AI platform is an intentional architectural choice. Unlike standard chatbot interfaces, Muse operates as a fully functional Linux virtual machine running in the cloud.
Google has launched its Gemini 3.8 Live update, introducing dynamic, real-time animated avatars that lip-sync and display facial expressions during voice interactions. Currently available to Enterprise customers, the feature supports 97 languages with zero visual drift.