Articles tagged with "AI-simulation"
Genie Sim 3.0 launches as AGIBOT expands open robotics simulation
AGIBOT has launched Genie Sim 3.0, an advanced open-source robot simulation platform unveiled at CES 2026, designed to accelerate the development and evaluation of embodied intelligence in robotics. The platform integrates tightly with NVIDIA Isaac Sim on NVIDIA Omniverse, creating a unified pipeline that combines digital asset creation, scene generation, data collection, automated evaluation, and physics-based simulation. Central to Genie Sim 3.0 is the Genie Sim Benchmark, a standardized evaluation system covering over 200 tasks across 100,000+ simulated scenarios, aiming to provide comprehensive capability profiles for robotic models rather than narrow metrics. A key feature of Genie Sim 3.0 is its extensive open-source dataset, which includes more than 10,000 hours of synthetic data reflecting real-world robot operations across multiple sensor modalities such as RGB-D, stereo vision, and whole-body kinematics. The platform incorporates automated data collection with teleoperation, scripted task execution, auto-annotation, and failure recovery to reduce time and
roboticsrobot-simulationembodied-intelligencedigital-twinsAI-simulationNVIDIA-Isaac-Simopen-source-roboticsDemonstrably Safe AI For Autonomous Driving - CleanTechnica
The article from CleanTechnica details Waymo’s approach to achieving demonstrably safe AI for autonomous driving, emphasizing safety as the foundational principle rather than an afterthought. Waymo has driven over 100 million fully autonomous miles, demonstrating a significant reduction in crashes with serious injuries compared to human drivers. Their AI ecosystem is built around a holistic strategy that integrates a Driver (the AI system controlling the vehicle), a Simulator for realistic closed-loop training and testing, and a Critic that evaluates performance and guides improvements. These components are unified by the Waymo Foundation Model, which enables continuous learning and safety validation at scale. The Waymo Foundation Model serves as the cornerstone of their AI system, combining the benefits of both end-to-end and modular AI architectures. It uses learned embeddings and structured representations (such as objects and road elements) to ensure correctness and safety during inference, efficient large-scale simulation, and strong feedback for training. The model employs a dual architecture known as Think Fast and Think Slow: a Sensor Fusion
robotautonomous-drivingAI-safetyWaymoautonomous-vehiclesAI-simulationAI-ecosystemVideo: UK humanoid robot walks just 48 hours after final assembly
UK-based robotics firm Humanoid has developed the HMND 01 Alpha, a humanoid robot that achieved stable walking just 48 hours after final assembly, marking a significant acceleration in humanoid robot development. This rapid progress was enabled by precise digital design and extensive simulation-led training using Nvidia Isaac Sim and Isaac Lab, which condensed nearly 19 months of conventional training into two days. The robot stands 179 cm tall, features 29 degrees of freedom, supports a 15 kg bimanual payload, and is equipped with multiple sensors and onboard computing power from Nvidia Jetson Orin AGX and an Intel i9 processor. Designed for applications in manufacturing, logistics, and home assistance, the Alpha robot can perform complex movements such as walking, running, squatting, and recovering from external forces up to 350 Newtons. It also has object manipulation capabilities and can communicate through displays, audio, and LEDs. The KinetIQ reasoning framework enables advanced planning and interaction. Humanoid’s CEO Artem
roboticshumanoid-robotindustrial-automationassistive-roboticsAI-simulationNvidia-Jetsonbipedal-robotThe titans of tech: Top 10 most powerful supercomputers of 2025
As of June 2025, the global landscape of supercomputing is dominated by exascale machines primarily based in the United States, with significant new entries from Europe and continued presence from cloud and industrial sectors. Leading the pack is El Capitan at Lawrence Livermore National Laboratory, boasting 1.742 exaFLOPS on the LINPACK benchmark and demonstrating balanced performance across scientific workloads. Following closely is Frontier at Oak Ridge National Laboratory, the first-ever exascale supercomputer, now ranked second with 1.353 exaFLOPS, maintaining its role in advanced scientific research. Aurora at Argonne National Laboratory rounds out the top three, achieving just over 1 exaFLOPS and designed to integrate simulation with AI-driven science applications. Europe's fastest system, Germany’s JUPITER Booster, marks a significant milestone by entering the top tier with 793.4 petaFLOPS, powered by NVIDIA Grace-Hopper superchips and InfiniBand networking
energymaterialssupercomputersexascale-computinghigh-performance-computingAMD-EPYCAI-simulationWhy Runway is eyeing the robotics industry for future revenue growth
Runway, a New York-based company known for its AI-powered video and photo generation tools built over the past seven years, is now targeting the robotics industry as a new avenue for revenue growth. The company’s advanced world models—large language models that simulate realistic versions of the real world—have attracted interest from robotics and self-driving car companies seeking scalable and cost-effective training simulations. Runway’s co-founder and CTO, Anastasis Germanidis, explained that while the company initially focused on creative and entertainment applications, inbound requests from robotics firms revealed broader use cases for their technology beyond entertainment. Robotics companies are leveraging Runway’s models to create highly specific training simulations that are difficult, costly, and time-consuming to replicate in real-world environments. These simulations allow for controlled testing of different actions and scenarios without altering other environmental variables, providing valuable insights into outcomes that physical testing cannot easily achieve. Rather than developing separate models for robotics and autonomous vehicles, Runway plans to fine-tune its existing models and is
roboticsAI-simulationself-driving-carsrobot-trainingvisual-generating-toolsrobotics-industrymachine-learningNew AI simulates 4 billion atoms to build carbon-neutral concrete
Researchers at USC Viterbi School of Engineering have developed Allegro-FM, an AI-powered simulation tool capable of modeling the behavior of over 4 billion atoms simultaneously with 97.5% efficiency. This represents a nearly 1,000-fold increase in scale compared to traditional models. Using the Aurora supercomputer at Argonne National Laboratory, Allegro-FM enables scientists to virtually test molecular compositions of concrete, accelerating the discovery of carbon-neutral, stronger, and fire-resistant formulations without costly lab experiments. The AI system can simulate complex atomic interactions across 89 chemical elements, making it versatile for applications beyond concrete, including carbon storage, battery chemistry, and biomedical devices. The key breakthrough is Allegro-FM’s ability to predict atomic interactions without relying on element-specific quantum mechanical equations, instead using machine learning trained on extensive datasets. This allows for highly accurate and efficient simulations of concrete’s mechanical and structural properties. The researchers found that incorporating CO₂ into concrete can trap carbon dioxide released during cement production,
materialsAI-simulationcarbon-neutral-concretesustainable-constructionatomic-modelingconcrete-durabilitycarbon-captureGenesis AI brings in $105M to build universal robotics foundation model - The Robot Report
Genesis AI, a physical AI research lab and robotics company, has emerged from stealth with $105 million in funding to develop a universal robotics foundation model (RFM) and a horizontal robotics platform. The company aims to advance "physical AI"—the intelligence enabling machines to perceive, understand, and interact with the real world—by leveraging digital AI foundations to create general-purpose robots with human-level intelligence. Founded by robotics Ph.D. Zhou Xian and former Mistral AI researcher Théophile Gervet, Genesis AI focuses on building a scalable data engine that unifies high-fidelity physics simulation, multimodal generative modeling, and large-scale real robot data collection to train robust, flexible, and cost-efficient robots. Physical labor accounts for an estimated $30 to $40 trillion of global GDP, yet over 95% remains unautomated due to limitations in current robotic systems, which are often narrow, brittle, and costly. Genesis AI seeks to overcome these challenges by generating rich synthetic data through
roboticsartificial-intelligencephysical-AIrobotics-foundation-modelautomationrobotics-platformAI-simulationMeet the Finalists: VivaTech’s 5 Most Visionary Startups of 2025
VivaTech 2025 has announced the five finalists for its inaugural Innovation of the Year award, selected from an initial pool of 30 visionary startups addressing global challenges with advanced, scalable technologies. The finalists represent diverse sectors including physics simulation, medical imaging, renewable energy, cancer diagnostics, and data center efficiency. They are BeyondMath, which offers a generative physics platform enabling engineers to run simulations 1000 times faster; Chipiron, developing an affordable, ultra-low magnetic field MRI for widespread early screening; Enerdrape, providing non-invasive geothermal panels for urban heating and cooling; Hua Tech International, creating an automated microfluidic platform for precise cancer cell analysis; and Lumisync, delivering a photonic oscillator to drastically reduce data center latency and energy use. These startups will present their pitches live on June 11th at VivaTech 2025, followed by a Q&A with expert judges from TechCrunch, 25madison, Daphni, and NVIDIA. The winner, announced on June 12th, will receive a free Startup Corner at VivaTech 2026 and entry into the TechCrunch Startup Battlefield 200. The awards ceremony will also honor winners of four other categories: the Female Founder Challenge, Africatech Awards (focusing on GreenTech, HealthTech, and E-commerce/FinTech), the Next Startupper Challenge for student entrepreneurs, and the new Tech For Change Award recognizing startups with positive social impact. VivaTech’s Innovation of the Year aims to spotlight startups with exceptional creativity and transformative potential across industries.
energyrenewable-energygeothermal-technologyAI-simulationphotonic-oscillatorsemiconductor-biochipmedical-imaging