SOURCE ARCHIVE · ORIGINAL LINKS INCLUDED
Google DeepMind unveils AlphaGenome Atlas, a free platform predicting the effects of all 9 billion possible single-letter DNA changes in the human genome to accelerate genetic research and disease understanding.
Google introduced agentic video understanding for Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite, reducing token use by up to 88% and costs by up to 66% while improving accuracy by up to 7%.
Google introduced Gemini Omni 1.1 Flash, enhancing AI-driven video generation with scene extension, frame interpolation, and 4K output for developers.
Google introduced Gemini 3.5 Transcribe, a new speech-to-text model designed for precise real-time transcription with reduced latency and improved accuracy across multiple languages.
Google DeepMind is collaborating with game developers to explore AI-driven gameplay innovations, building on decades of research from Atari to StarCraft and advancing projects like SIMA for interactive gaming agents.
Google DeepMind unveils SL2T, a sign-language-to-text model enabling Deaf users to dictate in ASL via Pixel 11’s Gboard and Live Transcribe, with additional languages planned.
Google DeepMind and collaborators open-sourced WeatherNext AI models, delivering cyclone forecasts with an extra day of lead time, equivalent to a decade of meteorological progress.
Google DeepMind has launched Gemini Robotics ER 2, a new model designed to serve as a high-level control system for robots, enabling real-time spatial reasoning, multi-step task planning, and multi-robot collaboration.
Google DeepMind unveils Gemini Robotics 2, an AI model enabling robots to perform whole-body tasks, advanced dexterity, and multi-robot collaboration, with local on-device operation and safety benchmarks.
Google DeepMind commits $40 million in AI tokens and cloud credits to support the U.S. Department of Energy’s Genesis Mission, aiming to double the pace of scientific discovery through expanded access to frontier AI...
Google introduced three new Gemini models—3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber—optimized for efficiency, cost, and agentic workflows, with pricing and performance details disclosed.
Google DeepMind introduces Gemini 3.5 Flash Cyber, a lightweight cybersecurity model designed to detect and patch software vulnerabilities efficiently, available initially to governments and trusted partners via a pilot...
Google DeepMind and Isomorphic Labs outlined a joint AI-driven strategy to strengthen global biosecurity through prevention, detection, and response measures, emphasizing responsible deployment and partnerships with...
Google DeepMind and Atal Innovation Mission launched ATL Saathi, a Gemini-powered assistant for 100 pilot schools, to support educators with AI-driven planning, training, and curriculum alignment in local languages.
Google DeepMind and A24 have launched a research partnership to develop new tools for filmmakers, combining AI expertise with creative input to shape future storytelling technologies.
Google introduced Nano Banana 2 Lite, a faster and more cost-efficient image generation model, and Gemini Omni Flash for high-quality video generation and editing. Both models are now available to developers and...
Google integrates computer use capabilities directly into its Gemini 3.5 Flash model, enabling developers to build cross-platform agents for automation tasks such as software testing and knowledge work.
Google DeepMind partners with the UK government to develop an AI prototype that aims to halve the time required to process householder planning applications, supporting the national target of 1.5 million new homes by...
Google DeepMind introduces an AI Control Roadmap to secure internal AI agents, addressing risks from misalignment or unexpected behavior despite advanced capabilities.
Google DeepMind released DiffusionGemma, an open 26B MoE model using diffusion for text generation, offering up to 4x faster inference on local GPUs compared to autoregressive models.
Google DeepMind and partners launch a $10M funding call to study safety risks in large-scale multi-agent AI systems, aiming to address emergent behaviors and systemic vulnerabilities before widespread deployment.
Google introduced Gemini 3.5 Live Translate, an audio model enabling near real-time speech-to-speech translation in over 70 languages with natural intonation and minimal delay.
Google introduces Gemma 4 12B, a multimodal AI model optimized for laptops, combining audio and visual inputs without separate encoders to reduce latency and memory use.
Google DeepMind launches a three-month accelerator program for 15 European robotics startups, offering mentorship, AI models, and technical support to integrate AI into their products.