50 articles found
While the tech world obsesses over humanoid robots and general-purpose AI, some of the most significant advances in physical AI are happening in the least glamorous place imaginable: construction sites. Recent partnerships and new platforms suggest the industry is finally ready to move beyond cautious pilot programs.
Pegasus Tech Ventures launched a $60 million corporate venture capital fund in partnership with CYBERDYNE Inc. to invest in startups developing robotics, physical AI, healthcare automation, and intelligent systems. The fund will focus on advancing CYBERDYNE's vision for human-cyber-physical space (HCPS) cybernics, combining humans, AI, robotics, and information systems to address global challenges like aging populations and workforce shortages.
Built Robotics is partnering with the University of Pennsylvania's Safe Autonomous Systems Lab (xLAB) to develop physical AI for construction sites, leveraging Built's 50,000+ hours of operational data and new data-collection robots. The collaboration aims to create a foundation model for safe human-machine coexistence on jobsites by systematically collecting edge case data and training AI models for improved human detection and hazard identification in complex construction environments.
The article features a story about how AI and data analytics are revolutionizing soccer through research at KU Leuven's Sports Analytics Lab, where computer scientists have uncovered hidden tactical patterns and challenged traditional assumptions about the game. The lab's work has influenced modern soccer strategy by applying machine learning and analytical techniques to uncover insights that are now being implemented on professional pitches.
Industry experts at the 2026 Robotics Summit & Expo discussed the current state of humanoid robot development, with Boston Dynamics' Alberto Rodriguez highlighting the company's strategy to build general-purpose Atlas robots for logistics and manufacturing applications, including plans to deploy 25,000 humanoids and reach 30,000 units per year by 2028. Panelists from Boston Dynamics, Agility Robotics, and other companies emphasized the challenges of designing bipedal robots that can work safely alongside human workers and adapt to diverse, one-of-a-kind jobs rather than single applications.
Google DeepMind is funding a $10 million research initiative to study the safety risks of multi-agent AI systems, concerned about potential dangers when millions of autonomous AI agents interact online. The funding, announced alongside Schmidt Sciences, the UK government's ARIA, and other partners, aims to establish a field of research around multi-agent safety, focusing on risks like coordinated scams, prompt injections, and cyberattacks that could emerge from agent deployments at scale.
Jesse Davis and his Sports Analytics Lab at KU Leuven use machine learning and advanced data analytics to uncover tactical insights in professional soccer, including research on optimal strategies like long throw-ins near the goal. The lab's open-source tools and research have influenced how top soccer clubs evaluate rosters, assess strategy efficiency, and discover hidden tactical patterns in the sport.
South Korea is deploying AI-powered companion dolls like Hyodol to combat elderly isolation and loneliness in a rapidly aging society facing record numbers of deaths among isolated seniors. The dolls, developed by startups including Hyodol and Mr. Mind, use conversational AI and ChatGPT to provide companionship, health reminders, and emotional support to seniors living alone, with around 14,500 units already in use across the country.
While the AI industry obsesses over chatbots and image generators, a quieter revolution is happening in embodied AI. Two major competitions this year—AGIBOT's World Challenge and AAMAS 2026's research awards—reveal where the real technical challenges lie: getting robots to actually perform useful tasks in messy, unpredictable environments.
A vision for the future of AI, focusing on access, safety, and shared prosperity as OpenAI works to ensure AGI benefits everyone.
AGIBOT hosted the World Challenge 2026 competition with 526 teams from 27 countries competing in embodied AI tasks using real robots and standardized benchmarks. The competition evaluated two tracks—Reasoning to Action (task understanding and execution) and World Model (physical world prediction)—with finalists completing challenges using the AGIBOT G2 humanoid robot, marking a shift toward real-robot validation over simulation-only evaluation. The competition also introduced a supermarket benchmark track for end-to-end mobile manipulation in realistic retail environments.
Georgia Tech researchers have developed COBALT, a smartphone app that enables remote operation of robot arms from anywhere in the world with just a Wi-Fi connection. The technology was presented at IEEE's International Conference on Robotics and Automation and aims to scale policy training data collection through crowdsourcing by leveraging the billions of smartphone users globally, while also expanding access to robotics education and enabling new gig economy applications for robotic task execution.
Federal courts are experiencing a surge in lawsuits filed by self-represented litigants, with research showing AI-generated content in legal filings rose from 1% in 2023 to 18% in 2026. While judges report that AI-drafted pleadings are often clearer and easier to understand than handwritten documents, self-represented litigants using AI still lose cases at higher rates than those with lawyers, and legal questions are emerging about chatbot liability and the duty of care for AI-generated legal advice.
Amazon's AGI group presents a new "audit-then-score" protocol for evaluating AI-generated research reports, addressing the challenge that traditional static benchmarks fail for complex AI evaluation tasks. The approach uses AI models to actively scrutinize and refine human-generated benchmarks, improving fact-checking accuracy from 60.8% to 90.9%, and reframes ground truth as an evolving process rather than a fixed dataset.
While the robotics industry obsesses over humanoid form factors and factory automation, a quiet revolution is happening in operating rooms. Incisionless surgery powered by AI and acoustic technology represents a fundamental reimagining of what surgical intervention can be—and it's already raising millions in funding.
MIT researchers have developed Masked Inverse Reinforcement Learning (Masked IRL), a system that uses large language models to help robots understand vague instructions and focus on relevant details from demonstrations. The approach reduces the need for extensive human demonstrations by 80% while enabling robots to safely complete manipulation tasks in warehouses, offices, and homes by automatically clarifying ambiguous prompts and identifying crucial environmental constraints.
Healthcare providers are increasingly deploying agentic AI systems to address workforce shortages and reduce administrative burden on clinicians. Hospital for Special Surgery (HSS) has implemented AI agents to automate insurance claims processing (handling 1,100 claims monthly with improved appeal success rates) and is now deploying conversational AI for patient scheduling and triage services, demonstrating how autonomous AI can augment clinical workflows while maintaining human oversight for complex cases.
A webinar panel discussion featuring industry leaders from RightHand Robotics, Locus Robotics, and Nomagic will explore advances in robotic case and item picking for order fulfillment, covering machine vision, AI-guided manipulation, and human-robot collaboration. Topics include addressing challenges in manipulating diverse shapes and materials, the role of AI training through data collection, and the future of human supervision in automated warehouse operations.
A long-overlooked organ may hold surprising clues to healthy aging and cancer survival. Researchers at Mass General Brigham used AI to analyze CT scans from tens of thousands of adults and found that...
A Java developer's deliberate sabotage of AI coding tools has ignited a conversation the industry desperately needs: what happens when the infrastructure we build on becomes weaponized against the very tools we're creating? The jqwik incident isn't just about one angry developer — it's a warning shot about trust, dependency, and the fragile ecosystem holding modern software together.
Warp is leveraging OpenAI's GPT-5.5 model to coordinate AI-powered coding agents across different development environments, combining local, cloud, and open-source workflows. The company is betting on using advanced language models to automate and streamline software development processes.
The 2026 Robotics Summit & Expo in Boston will convene over 5,000 robotics experts and 200+ exhibitors across two days, featuring 50+ sessions and 70+ speakers covering AI, humanoids, healthcare, and logistics. Key sessions include keynotes on robot autonomy, the state of humanoids (with speakers from Boston Dynamics, Agility, and others), and presentations from major companies like Tesla, Toyota Research Institute, and Universal Robots.
A University of Vaasa doctoral study finds that employees who view generative AI tools like ChatGPT as collaborative partners rather than threats show higher engagement and adaptability. The research emphasizes that workers won't be replaced by AI itself, but by those who learn to use AI effectively, and proposes an eight-step framework for organizations to strategically implement AI while managing trust, ethics, and governance.
Researchers at Helmholtz Munich developed MouseMapper, an AI-powered system that maps entire mouse bodies at cellular-level detail using deep learning foundation models and whole-body imaging. The system discovered that obesity causes previously unknown damage to facial sensory nerves and widespread inflammation, with similar molecular patterns identified in human tissue, published in Nature.
OpenAI has been named a Leader in Gartner's 2026 Magic Quadrant for Enterprise AI Coding Agents, with its Codex model recognized for innovation and enterprise-scale deployment capabilities. This recognition highlights OpenAI's position in the AI-assisted software development market.
Biologists use Co-Scientist to find novel factors that successfully rejuvenate human cells.
Calico Life Sciences uses Co-Scientist to connect scattered findings and generate new leads in aging research.
Researchers at Jülich Supercomputing Centre and NVIDIA achieved a record-breaking milestone by simulating a 50-qubit quantum computer using JUPITER, Europe's first exascale supercomputer, surpassing the previous 48-qubit record. The breakthrough demonstrates how advanced supercomputing enables quantum research validation and algorithm development, with the simulation requiring 2 petabytes of memory and managing over 2 quadrillion complex numerical values across thousands of computing nodes.
Co-Scientist is a multi-agent AI system designed to accelerate scientific research by synthesizing biomedical literature, proposing testable hypotheses, and identifying promising research directions. The platform has been deployed across multiple research institutions to advance work on liver fibrosis, ALS, cellular aging, infectious disease mechanisms, and aging biology, with researchers reporting significant time savings and improved research productivity.
Cognex launched the In-Sight 3900, an AI-powered vision system for factory automation that delivers 4X faster processing and up to 25MP resolution while running deterministic, real-time inspections at full production speed. The system combines edge AI, advanced AI, and rule-based vision tools with Qualcomm Dragonwing processors to eliminate traditional tradeoffs between inspection depth and line speed in manufacturing applications like packaging, automotive, and electronics.
Colin Angle, co-founder of iRobot, has launched Familiar Machines & Magic, a startup developing consumer companion robots called Familiars designed to form emotionally intelligent, long-term relationships with users. The company, which emerged from stealth, combines Angle's 30+ years of consumer robotics experience with a team including veterans from iRobot, Amazon, Disney Research, MIT, and Boston Dynamics to build a platform for artificial life.
OpenAI and PwC are collaborating to develop AI agents that automate financial workflows for CFOs, including procurement, payments, treasury, and accounting processes. The partnership leverages OpenAI's internal finance operations as a testing ground, where AI tools have already demonstrated significant efficiency gains like processing 5x more contracts with the same team size.
Colin Angle, founder of iRobot, launched a new startup called Familiar Machines & Magic with a companion robot designed to perform "emotional work" and reduce loneliness through physical presence and interaction. The robot, called a "familiar," is a small quadruped with 23 degrees of freedom, fuzzy exterior, edge AI processing, and communicates through motion and behavior rather than screens or speech, aiming to integrate into daily life in ways existing companion robots have failed to achieve.
Teradyne Robotics reported $91 million in Q1 2026 revenue, marking its fourth consecutive quarter of growth, with strength in e-commerce, electronics manufacturing, and semiconductor end markets including AI data centers. The company's robotics division, which includes Universal Robots cobots and Mobile Industrial Robots, saw AI revenue reach approximately 15% of quarterly sales, and the company is leveraging robotic-assisted assembly and data center operations as part of its wafer-to-AI strategy.
Software engineer Chris Nager created a playable Doom application that runs within AI chatbots like ChatGPT and Claude by leveraging Model Context Protocol (MCP), an open-source standard that allows LLMs to connect with external tools and data sources. Users can launch the game by typing "play Doom" directly in the chatbot interface, with fallback URL access if inline launching isn't supported by the client.
OpenAI is scaling its Stargate project to build compute infrastructure designed to power artificial general intelligence (AGI) development, expanding data center capacity to meet increasing demands for AI training and deployment.
SquareMind raised $18 million to commercialize Swan, a robotic platform that uses AI and robotics to automate full-body dermoscopic skin imaging for melanoma detection and tracking. The funding, led by Sonder Capital, will support the company's preparation for U.S. and European market launch, addressing dermatology's capacity constraints and improving clinical documentation efficiency.
OpenAI has released Symphony, an open-source specification for orchestrating AI coding agents that uses project management tools like Linear as a control plane. The system allows multiple coding agents to autonomously pull tasks from an issue tracker and work continuously with human review, achieving a 500% increase in pull request throughput by eliminating the bottleneck of managing interactive coding sessions.
The article explores practical applications of Codex, OpenAI's code generation AI model, demonstrating how it can be used to automate tasks, generate code outputs, and integrate workflows across various tools and files. It presents 10 use cases for leveraging Codex in everyday work scenarios.
If you've been to an eye doctor and had an image taken of the inside of your eye, chances are good it was done with optical coherence tomography (OCT)-a technology invented by clinician-scientist Davi...
The article examines how criminals are leveraging generative AI and large language models to enhance cyberattacks, including phishing, deepfakes, malware development, and vulnerability detection at scale. While AI tools lower barriers for attackers and amplify attack volumes, defensive AI systems—such as Microsoft's threat detection processing over 100 trillion daily signals—are also being deployed to counter these threats, though future sophistication remains uncertain.
GRASP is a new gradient-based planning algorithm for learned world models that enables practical long-horizon planning by lifting trajectories into virtual states, adding stochasticity for exploration, and reshaping gradients to avoid optimization brittleness. The research addresses fundamental challenges in using large-scale world models for control and planning, tackling issues like ill-conditioned optimization and high-dimensional latent space failure modes that make gradient-based planning fragile at longer horizons.
Hyatt has deployed ChatGPT Enterprise across its global workforce, leveraging OpenAI's AI models to enhance employee productivity and operational efficiency. The hotel chain is using the technology to improve guest experiences and streamline internal operations through enterprise-wide AI adoption.
AGIBOT announced the deployment of its G2 semi-humanoid robots into consumer electronics manufacturing at Longcheer Technology, marking a transition from laboratory testing to large-scale industrial production. The robots achieve 310 units per hour throughput with over 99% success rate and 24/7 autonomous operation, performing precision loading, unloading, and quality control tasks on tablet production lines. Longcheer plans to deploy 100 AGIBOT G2 robots by Q3, demonstrating the commercial viability of embodied AI systems in flexible manufacturing environments.
Perplexity has launched Personal Computer, an AI assistant for Mac that uses multi-model orchestration to act as a computer use agent capable of managing files, apps, and workflows on behalf of users. The software, available to Max subscribers, can handle tasks like organizing files and managing to-do lists while maintaining audit trails and reversible actions for user control.
Antioch, a startup founded in May 2025 by former Tesla Autopilot and Google DeepMind engineers, raised $8.5 million to develop cloud-based simulation tools for autonomous robot and system validation. The platform aims to eliminate expensive physical testing by leveraging simulation engines like NVIDIA Isaac Sim and generative AI techniques, allowing robotics teams to conduct testing and validation at software speed rather than hardware speed.
Amazon's Generative AI Innovation Center developed customized Amazon Nova language models fine-tuned for molecular-property prediction in drug discovery, demonstrating that a single optimized LLM can match the accuracy of multiple specialized graph neural networks while providing improved usability and reasoning capabilities for medicinal chemists. The approach combines supervised and reinforcement fine-tuning to create a unified AI assistant that simplifies workflows and accelerates drug discovery by enabling chemists to query multiple molecular properties in one interaction rather than managing separate models.
Adobe has launched Firefly AI Assistant, a prompt-based AI tool that orchestrates complex multi-step creative workflows across Photoshop, Premiere, Lightroom, and other Adobe apps. The assistant allows users to describe desired outcomes and automatically executes the necessary steps while maintaining user control and Adobe's native file formats, positioning itself against AI competitors by leveraging Adobe's professional creative tools.
AGIBOT launched Genie Studio Agent, a zero-code application platform designed to simplify robot development and deployment at scale without requiring coding expertise. The platform features no-code workflow orchestration, simulation-first deployment, real-world reinforcement learning optimization, and comprehensive monitoring—addressing the critical challenge of efficiently deploying embodied AI robots in real-world environments.
AGIBOT announced Genie Envisioner 2.0, a world simulator that evolves beyond traditional world models to create fully interactive, physics-consistent environments for training embodied AI systems. The system enables robots to be trained and optimized at scale in synthetic environments through action-conditioned world modeling, addressing key limitations of real-world trial-and-error approaches and representing a potential inflection point for scaling embodied intelligence.
