Google's new AI gives robots a smarter brain, and genomes a sharper decoder

Google DeepMind announced two major AI advances today: Gemini Robotics 2.5 and AlphaGenome 2. Gemini Robotics 2.5 is a vision-language-action model designed to be the 'AI brain' for robots, significantly improving dexterity, reliability, and safety for real-world use. In tests, it showed robots could perform delicate tasks with everyday tools and successfully packed automotive parts in a real-world proof-of-concept with a 96% success rate. The model works across various robot hardware, including Apptronik's Apollo 2 humanoids. Meanwhile, AlphaGenome 2 is a biological AI model that predicts how DNA sequences and their variations affect human cell types, aiming to accelerate disease research and drug development.

For engineers, Gemini Robotics 2.5 represents a tangible step toward deploying robots outside controlled lab environments, offering faster learning and higher task success rates. The partnership with companies like Mercedes-Benz and Apptronik shows a clear path from research to industrial application. For the broader world, these dual releases highlight AI's expanding frontier: one model aims to automate physical tasks alongside humans, while the other delves into the fundamental biological code of life itself. The takeaway is that AI is no longer just about generating text or images—it's increasingly about interacting with and understanding the physical and biological world.