Industry Briefing
A single destination for timely, editor-curated robotics news from around the world.
Unitree R1 Hits AliExpress: The TIME "Best Invention" Goes Global
Unitree Robotics has launched its R1 humanoid robot on AliExpress, a significant milestone in the company's efforts to expand internationally. The R1 is priced starting at $6,800, making advanced bipedal robotics more accessible to a wider audience. This development comes on the heels of a strategic partnership with Alibaba, aimed at promoting the democratization of robotics technology. The listing on AliExpress not only enhances Unitree's global presence but also reflects a growing interest in humanoid robots for various applications.
Unitree Robotics AliExpress R1
Unitree R1 and Figure 03 Named to TIME's Best Inventions of 2025, Highlighting Humanoid Momentum
TIME Magazine has included two humanoid robots, the Unitree R1 and Figure 03, in its annual Best Inventions list, highlighting their significance in the evolving field of bipedal robotics. The recognition comes as the demand for affordable and domestically-oriented robotic solutions increases, showcasing the technology's potential to integrate into everyday life. The Unitree R1 is noted for its cost-effectiveness, while the Figure 03 is designed with a focus on home use. This acknowledgment by a prominent publication underscores the growing relevance and acceptance of robotics in mainstream society, reflecting advancements in technology and consumer interest.
Unitree Robotics Figure Figure-03 time-magazine R1 robotics
Unitree CEO Predicts R1 Will Be a Bestseller, Teases Larger 1.8-Meter Humanoid
Unitree Robotics CEO Wang Xingxing announced that the company’s new R1 robot, designed to be affordable, is expected to lead global sales in the upcoming year. This projection reflects the growing demand for cost-effective robotic solutions in various sectors. Additionally, Wang revealed plans to introduce a larger, more agile humanoid robot, measuring 1.8 meters, before the end of this year. This strategic move aims to enhance the company’s product lineup and cater to a wider range of applications in robotics. The launch of these innovations underscores Unitree Robotics' commitment to advancing technology and maintaining a competitive edge in the rapidly evolving robotics market.
H1 Unitree Robotics R1
Video: KEENON Robotics Enters Humanoid Field with XMAN-R1, Expands Autonomous Cleaning Lineup
KEENON Robotics has introduced the XMAN-R1, its inaugural humanoid robot aimed at enhancing human interaction and collaboration with its existing line of commercial robots. This launch took place recently as part of the company's efforts to innovate within the service robotics sector. In addition to the XMAN-R1, KEENON also unveiled three new cleaning robots—KLEENBOT C20, C40, and C55—each designed to cater to various cleaning needs across different environments. This expansion of their product lineup underscores KEENON's commitment to providing a comprehensive range of automated solutions, enhancing efficiency and service capabilities in commercial settings.
Keenon XMAN-R1
LaST-R1: New Physical Reasoning Paradigm Achieves 99.9% Success Rate on LIBERO Benchmark
A collaborative research effort involving Simplexity Robotics, Peking University, and the Chinese University of Hong Kong (CUHK) has introduced LaST-R1, an innovative embodied AI paradigm. This new technology has demonstrated a remarkable 99.9% success rate on the LIBERO benchmark, surpassing the previous benchmark, π0.5, by 22.5% in real-world applications. The research highlights significant advancements in the field of artificial intelligence, showcasing the potential for enhanced performance in practical tasks. The findings were released in October 2023, marking a notable achievement in the ongoing development of AI systems.
AI
Peking University Team Proposes LaST-R1 Framework, Enabling Robots to 'Think and Act' Simultaneously!
Researchers from Peking University and the Chinese University of Hong Kong have unveiled the LaST-R1 framework, a groundbreaking integration of latent reasoning into reinforcement learning. This development, announced recently, aims to enhance the decision-making capabilities of robots by enabling them to not only execute actions but also comprehend and adapt to various physical states. The framework is designed to improve robots' performance in real-world scenarios, addressing the limitations of traditional reinforcement learning methods. By incorporating latent reasoning, the LaST-R1 framework represents a significant advancement in the field of robotics, potentially leading to more intelligent and responsive machines.
Reinforcement Learning Robot Perception AI Frameworks Physical Reasoning
Silicon Valley stunned as China’s DeepSeek-R1 surpasses US AI giants, undercuts costs
Tensions escalated in Silicon Valley this week following the launch of DeepSeek's R1 model, a Chinese AI technology that has surpassed leading U.S. companies such as OpenAI, Meta, and Anthropic in third-party performance benchmarks. The unease began earlier with the release of DeepSeek v3, which already outperformed Meta’s Llama 4. The R1 model, unveiled on January 20, has been noted for its significant improvements in model inference capabilities, raising concerns among American tech firms about competition and innovation in the rapidly evolving AI landscape.
News Feed
Unitree Expands R1 Lineup with Dual-Arm Modular Platform Starting at $4,290
Unitree Robotics has introduced a new series of dual-arm humanoid robots equipped with both fixed and wheeled bases, aiming to provide a cost-effective solution for the development of complex tasks. This launch, which took place recently, highlights the company's commitment to making advanced robotics more accessible to a wider audience. By offering these versatile robotic platforms, Unitree Robotics seeks to facilitate innovation in various industries, enabling users to explore automation solutions without the significant financial investment typically associated with such technology. The dual-arm design enhances the robots' ability to perform intricate operations, making them suitable for a range of applications from research to practical deployment in commercial settings.
Unitree Robotics China R1-D
NASA Picks Pierce Aerospace for Bay Area Remote ID Sensor Network
Pierce Aerospace has been selected by NASA to develop a Remote ID network throughout Silicon Valley and the San Francisco Bay Area. This initiative is part of NASA's ongoing research into drones and air taxi technologies, specifically at the Ames Research Center. The project will involve the deployment of YR1 and YR2S sensors, facilitated by Metis Technology Solutions, an Indiana-based firm. The collaboration aims to enhance the safety and efficiency of unmanned aerial systems in urban environments, addressing the growing need for regulatory frameworks as drone usage expands. The timeline for the deployment and operational details has not been disclosed, but the network is expected to play a crucial role in advancing aerial mobility solutions in the region.
Drone News Drone News Feeds infrastructure Networking News Sensors
Unitree Unveils H2, a Full-Sized Humanoid Successor to the H1
Unitree has unveiled its latest flagship humanoid robot, the H2, which stands at an impressive 1.8 meters tall. This new model represents a significant advancement over the company's previous affordable offering, the R1, and aligns with Unitree's strategy to cater to different segments of the market. The announcement highlights the company's commitment to innovation and development in robotics, showcasing the H2 as a more sophisticated option for consumers and businesses alike. The launch reflects Unitree's ongoing efforts to expand its product lineup and meet the evolving demands of the robotics industry.
H1 Unitree Robotics H2 R1
Swancor Advanced Materials to Launch World's First Shape-Shifting Personal Robot at WAIC
Swancor Advanced Materials is set to unveil the Quester1, the world's first shape-shifting personal robot, at the 2026 World Artificial Intelligence Conference (WAIC) on July 17. This innovative robot can automatically switch between bipedal and quadrupedal modes without manual disassembly, showcasing a unique 'transformable cross-modal integration architecture' that enables smooth transitions. The Quester1 is designed for various environments, with its bipedal form ideal for flat surfaces and intricate tasks, while the quadrupedal mode excels in navigating stairs and rough terrain. CEO Tian Hua emphasized that this product aims to redefine personal robotics for consumer markets, marking Swancor's first foray into the consumer robotics sector. As the Quester1 prepares for its debut, it highlights the rapid development of transformable robotics in China. The WAIC will feature over 1,100 companies and more than 300 product launches, indicating a significant shift towards embodied intelligence in robotics. No further timeline was disclosed at the time of publication.
Personal Robots Shape-Shifting Technology AI Robotics Consumer Robotics
KUKA Addresses Upcoming ISO 10218:2025 Robot Safety Standards and Market Implications
KUKA, a leading robotics provider for the automotive sector, is responding to increasing inquiries regarding the updated ISO 10218 safety standard. This standard, which will become mandatory under the new European Machinery Regulation by 2027, is crucial for CE-marked industrial robots. The transition may be delayed due to weak manufacturing demand and cost pressures on SMEs, raising concerns about supplier readiness. The significance of the ISO 10218:2025 update lies in its potential to enhance the competitive position of established vendors while posing market access risks for less prepared suppliers. The current Machinery Directive will be replaced by the Machinery Regulation EU 2023/1230, effective January 20, 2027. However, the timeline for the formal listing of the new standard in the Official Journal of the European Union remains uncertain, which could impact compliance timelines for manufacturers. In the U.S., the updated R15.06-2025 safety standard, released in September 2025, aligns with international standards but remains voluntary. Despite this flexibility, major customers often require compliance, and non-adherence can lead to penalties. As the European market faces subdued demand, suppliers that effectively manage the transition to the new safety standards will be better positioned to meet the needs of a safety-conscious customer base as demand recovers.
Manufacturing News Opinion Regulatory & Compliance Safety & Security association for advancing automation
AI Models Overthink Problems—and It’s a Security Risk
Large language models (LLMs) that can think through problems step-by-step have significantly increased the scope of tasks that AI can tackle. But new research suggests these reasoning capabilities also introduce a critical vulnerability that could allow attackers to slow these systems to a crawl.While earlier generations of LLMs would immediately produce a response to a user’s request, today’s most advanced models generate an internal monologue where they break down the problem into steps and reason about the best way to tackle it before providing an answer. This has allowed AI to tackle increasingly complex problems, particularly in areas like coding and math.However, previous research has shown that these models are susceptible to sometimes producing excessively long streams of reasoning that do little to boost performance, a phenomenon known as “overthinking.” In research presented this week at the International Conference on Machine Learning 2026 in Seoul, researchers from Zhejiang University and e-commerce giant Alibaba in China demonstrate that they can deliberately induce overthinking by subjecting models to logically inconsistent prompts. The result is a form of denial-of-service attack on commercial AI models.Evolutionary Prompt Attack on LLMsThe team has developed an evolutionary algorithm that corrupts the logical structure of prompts, causing models to spiral into overthinking as they attempt to reason through fundamentally unsolvable problems. Generating longer responses costs more and increases the load on a model provider’s servers, so if done at scale, the researchers say, this could significantly degrade the experience of legitimate users. The attack was effective against reasoning models from leading AI companies including DeepSeek-R1, Alibaba’s Qwen3-Thinking, OpenAI’s GPT-o3, and Google’s Gemini 2.5 Flash and resulted in outputs up to 26 times as long as standard responses on a standard math benchmark.“Across multiple datasets and reasoning models, our method substantially amplifies the output length,” Wei Cao, a masters student at Zhejiang University, wrote in an email to IEEE Spectrum. “Our results suggest that overthinking is not an isolated phenomenon specific to individual models, but rather a shared vulnerability among modern reasoning models.”The team’s approach builds on previous research from another group of researchers that showed reasoning models tend to overthink when faced with a question in which a key premise has been removed—such as asking how far someone who walks ten miles a day covers in total without specifying how many days they walked for. Rather than identifying that the problem is unsolvable, models often engage in extended but ultimately fruitless reasoning loops in an attempt to answer the question.Taking the idea a step further, the authors took 940 problems from three math benchmark datasets and used an LLM to break down their logical structure into a set of premises and a final question. The genetic algorithm then jumbled these up using a variety of “mutations,” including swapping premises between problems, adding extra premises to problems, deleting existing premises from problems, and swapping the final questions between two sets of premises.After each round of mutations, the problems are scored on how many words they cause a target model to output and also whether they increase the frequency of specific linguistic markers of overthinking—words like “but,” “wait,” “maybe,” or “alternatively.” The problems that scored highest on both measures are retained and the remaining ones are jumbled up again, and this process is repeated for five generations. Crucially, the approach doesn’t require access to the internals of a model and can generate malicious prompts by simply querying the target, which makes it possible to attack closed-source commercial services, says Cao.Overthinking Vulnerability in AI ModelsThe researchers found that the approach consistently led to outputs several times longer than those generated by the unmodified questions for the reasoning models they tested it on. The biggest jump came from DeepSeek-R1 on the MATH dataset, which is made up of problems from high school math competitions, where the maximum output was 26.1 times as long as the longest response the model provided to unaltered questions. While the main thrust of the research was focused on math problems, the authors also tested it on coding, scientific reasoning, and dialogue challenges, and observed significant jumps in output length in all three.One challenge for the approach is that developing the malicious prompts requires repeated queries to expensive reasoning models, which Cao admitted could limit its cost-effectiveness. However, the researchers also demonstrated that when they used a smaller, cheaper model to generate the malicious prompts they were still able to induce the target models to produce outputs several times longer than normal. This ability to transfer malicious prompts between models significantly increases the attack’s feasibility, Cao wrote.However, he pointed out that the goal of the research is not to develop a practical DoS attack on reasoning models. Factors like the providers’ pricing model, rate limiting policies, context window size, and existing defenses could all impact how effective the approach is. The intention is instead to highlight these models’ vulnerability to logically inconsistent prompts so that providers can attempt to mitigate the problem.“Our objective is not to demonstrate that large-scale attacks can be launched at negligible cost, but rather to establish that this attack surface exists,” he wrote. “Our results indicate that the vulnerability represents a realistic security concern.”
Llms Artificial-intelligence Denial-of-service Cybersecurity
Unbounded Dynamics Launches MWA™ Model, Tops Global Embodied Intelligence Rankings
Unbounded Dynamics has introduced the MWA™ model, marking a significant advancement in embodied intelligence as it becomes the first long-sequence bidirectional physical causal chain. This innovative model has secured the top ranking in the RoboCasa GR1 TableTop competition, outperforming industry giants such as NVIDIA and Xiaopeng. By incorporating a distinctive 'hidden space world model' alongside reinforcement learning, the MWA™ model equips robots with the ability to comprehend physical laws and implement accurate strategies in intricate real-world environments. This development highlights Unbounded Dynamics' commitment to pushing the boundaries of robotic intelligence and performance.
Embodied Intelligence Reinforcement Learning Robotics Artificial Intelligence
Bear Robotics acquires Kinisi Robotics to boost its physical AI capabilities
Bear Robotics has announced the acquisition of Kinisi Robotics, a move aimed at enhancing its physical AI capabilities. The integration will involve Kinisi's KR1 humanoid robot and the expertise of its engineering team based in Bristol. This strategic acquisition reflects Bear Robotics' commitment to advancing its technology in the robotics sector, particularly in the realm of physical AI. The deal is expected to bolster Bear Robotics' offerings and position the company for future growth in the competitive landscape of robotic solutions.
Artificial Intelligence Artificial Intelligence / Cognition Autonomous Mobile Robots (AMRs) Design / Development Financial Humanoids
Vention and Teradyne Robotics Collaborate on Digital Twin Creation Platform Optimized for UR Robotic Cells
At Automate 2026, live demonstrations highlighted the innovative integration of Vention's Rapid Operator AI with NVIDIA's Isaac open robotics platform. The event showcased advanced welding applications facilitated by Universal Robots' UR12e and UR20 collaborative robot models. These demonstrations, taking place in a cutting-edge robotics environment, aimed to illustrate the potential of AI-driven automation in enhancing manufacturing processes. By combining these technologies, the event underscored the growing importance of robotics in streamlining operations and improving efficiency in various industrial applications.
The Rivian R2 is too much fun to let drive itself
Rivian, the electric vehicle manufacturer known for its R1S SUV and R1T pickup, is shifting its focus towards mainstream success with the upcoming launch of its R2 model in 2027. While the company has garnered significant media attention and a dedicated following for its high-end vehicles, it now faces the challenge of appealing to a broader audience of electric vehicle buyers. As Rivian continues to explore the potential of robotaxis and autonomous driving, it recognizes the importance of winning over consumers who prefer traditional human-driven vehicles. The R2 is seen as a pivotal step in this strategy, aimed at making electric vehicles more accessible and appealing to the general public.
Autonomous Cars Electric Cars Reviews Tech Transportation
ROKAE Robotics Expands Humanoid Series with New Models
ROKAE Robotics has unveiled its latest additions to the AR humanoid force-controlled arm series, introducing the AR3 and AR10 models. This launch, which builds upon the established AR5 platform, aims to meet the increasing demand for accurate, human-like manipulation in industrial applications of embodied intelligence. By expanding its product lineup, ROKAE Robotics seeks to enhance its offerings and provide comprehensive solutions for industries requiring advanced robotic capabilities.
Leading Company in Wheeled Embodied Intelligence Robots Launches New Product
Xinghai Map, a frontrunner in the wheeled robotics industry, has announced the launch of its upgraded R1 series, set to debut in 2026. This new series focuses on enhancing hardware precision and reliability, with the goal of setting new benchmarks in the field of embodied intelligent robotics. As the technology transitions from laboratory settings to practical applications, the R1 series is expected to play a significant role in advancing the capabilities of robotic systems in various real-world scenarios.
Embodied Intelligence Robotics Hardware Development AI
Beyond the Backflip: Zachary Jackowski on Why Generalization is the Final Frontier for Atlas
In a recent interview, Zachary Jackowski, the lead engineer for Boston Dynamics' Atlas robot, announced the retirement of the research-grade R1 model. He explained that the company is shifting to a unified "Thor" compute module to streamline operations. Jackowski emphasized that despite advancements in technology, the Atlas robot remains too advanced for typical household use, indicating that its capabilities exceed what is necessary for everyday applications. This transition reflects Boston Dynamics' ongoing commitment to innovation while addressing practical considerations for consumer robotics.
hyundai US Atlas Boston Dynamics
Unitree Scales Global Ambitions with Reported Alibaba Strategic Partnership
Unitree Robotics is set to collaborate with Alibaba’s AliExpress to introduce its R1 humanoid robot to international markets, a move that signals a major step towards making advanced robotics more accessible to consumers. This partnership comes as Unitree prepares for its anticipated initial public offering, which is projected to raise $580 million. The launch aims to capitalize on the growing demand for innovative technology solutions in everyday life, positioning the R1 as a viable option for consumers seeking cutting-edge robotics. The initiative is expected to enhance the visibility and availability of humanoid robots, paving the way for broader adoption in various sectors.
Unitree Robotics Market China
Kinisi Robotics Deploys Wheeled Humanoid for Glass Sorting
A startup based in New York City and Bristol has launched its first practical application of technology by deploying its wheeled KR1 robot to sort glass bottles at a facility in the UK. This marks a significant milestone for the company as it transitions from laboratory development to real-world operations. The initiative aims to enhance recycling efficiency and address the growing need for sustainable waste management solutions. By utilizing advanced robotics, the startup seeks to streamline the sorting process, reduce human labor, and improve overall productivity in recycling facilities. The KR1 robot's deployment is part of a broader effort to innovate within the recycling industry and contribute to environmental sustainability.
Kinisi
Alibaba launches preorders for its first in-house AI glasses, featuring AR navigation and visual payment integration across its ecosystem
Alibaba has launched preorders for its inaugural self-developed Quark AI Glasses, priced at ¥3,699 ($510) for members and ¥3,999 ($550) for non-members. The preorder phase is currently active on Tmall, with deliveries anticipated to begin in early December. These innovative glasses are equipped with Qualcomm’s AR1 and BES2800 dual flagship chips, allowing them to seamlessly integrate with Alibaba’s ecosystem services and provide augmented reality experiences. This move marks Alibaba's entry into the growing market for smart eyewear, aiming to enhance user interaction with digital content.
News Feed
Unitree unveils H2 humanoid robot with lifelike design
Chinese robotics company Unitree has introduced its latest humanoid robot, the Unitree H2, which stands 180 cm tall and weighs 70 kg. The unveiling took place recently, showcasing the robot's more humanlike features, including a bionic face and clothing, aimed at enhancing its realism during a demonstration. The H2 boasts 31 joints, representing a 19% increase from its predecessor, the R1 model, which allows for improved mobility and functionality. This advancement reflects Unitree's commitment to pushing the boundaries of robotics technology and creating more lifelike machines.
News Feed
Anti-Drone 5.56mm Rifle Rounds That Break Into Multiple Projectiles Sought By Marines
Marines are seeking specialized 5.56mm rifle rounds designed to enhance their capability to neutralize small unmanned aerial vehicles (UAVs). This initiative aims to equip Marines with ammunition that breaks into multiple projectiles upon firing, thereby increasing the likelihood of successfully targeting and disabling these uncrewed attackers. The development of these anti-drone rounds reflects a growing recognition of the evolving threats posed by drone technology on the battlefield. By improving their firepower against such aerial threats, the Marines hope to bolster their operational effectiveness in various combat scenarios.
Land Assault Rifles Drones M16/AR15/M4 Pattern News & Features SeaRobotics needs a service framework.
RSF defines a common language for robot service capability, lifecycle operations, certification pathways, and service-provider networks.
