Agibot’s A2 Humanoid Robot Sets World Record with 106-km Walk
A Chinese-developed humanoid robot, the A2 from Shanghai-based Agibot, has set a new Guinness World Record for the longest continuous walk by a humanoid robot. The robot completed a 106.286-kilometer journey from Suzhou to Shanghai without shutting down. The trek, which ran from November 10th to November 13th, showcased the maturity of the robot’s hardware, advanced balance algorithms, and exceptional endurance. The A2 was powered by a rapid hot-swap battery system, enabling it to remain active for the entire duration. This record-breaking walk marks a significant milestone for the large-scale commercial deployment of humanoid robots in various industries.
Sources:
DeepRoute.ai Scales Autonomous Driving, Equipping 200,000 Vehicles by 2025
Autonomous technology company DeepRoute.ai announced it is on track to equip over 200,000 vehicles with its advanced driver-assist and autonomous driving platform by the end of 2025. At the Guangzhou Auto Show, the company revealed it had already captured approximately 40% of China’s third-party Navigate on Autopilot (NOA) market in October alone. This major production milestone signals a strategic shift for DeepRoute.ai from R&D to large-scale commercial deployment. The company has ambitious plans for 2026, aiming to reach nearly one million cumulative vehicle integrations. This growth is supported by a unified technical architecture that serves consumer vehicles, robotaxi services, and its general-purpose autonomy framework.
Sources:
Google Launches Gemini 3: A New Generative AI Model Outperforming ChatGPT
Google has officially launched its next-generation generative AI model, Gemini 3, which is being integrated into core products like Google Search. The new model has demonstrated superior performance in independent tests, achieving the highest score ever in the Humanity’s Last Exam analysis with a 37.5% grade, surpassing ChatGPT 5.1’s 26.5%. Gemini 3 is engineered to handle complex tasks such as coding, image recognition, and content creation with fewer prompts. An enhanced version, Gemini 3 Deep Think, further improves logic and reasoning capabilities, scoring 41% on the same exam. While Gemini 3 leads in performance benchmarks, ChatGPT currently maintains a larger market share in monthly visitor traffic.
Sources:
QInscribe’s Generative AI Slashes Medical Report Drafting Time by 90%
QInscribe, a new medical writing service provider, has launched a generative AI solution that dramatically accelerates the creation of regulatory and scientific documents. The platform’s advanced AI-supported workflows can reduce the time to generate a draft Clinical Study Report (CSR) from a traditional 50-100 hours to as little as 5 hours—a 90% reduction. QInscribe aims to help pharmaceutical and biotechnology companies streamline regulatory submissions and improve document consistency by blending AI efficiency with the expertise of experienced medical writers. This innovative approach is designed to manage high volumes and tight deadlines for critical medical documentation.
Sources:
CNCF Announces Retirement of Kubernetes Ingress NGINX Controller
The Cloud Native Computing Foundation’s (CNCF) Kubernetes SIG Network and Security Response Committee have announced the upcoming retirement of the widely used Ingress NGINX controller. The project will receive best-effort maintenance until March 2026, after which no further releases, bug fixes, or security vulnerability updates will be provided. While existing deployments of Ingress NGINX will continue to function and installation artifacts will remain available, the Kubernetes team strongly recommends that users migrate to an alternative Ingress controller. The official successor is the Gateway API, which is positioned as the modern, more extensible replacement for Ingress. This decision was made to prioritize user safety after efforts to secure additional support for the project were exhausted.
Sources:
TRUMPF Taps Quantum Computing to Revolutionize Laser Development
High-tech company TRUMPF has partnered with the Fraunhofer Institute for Laser Technology ILT and the Dahlem Center for Complex Quantum Systems at Freie Universität Berlin to explore quantum algorithms for laser physics research. This collaboration aims to determine if quantum computers can more accurately simulate the complex quantum mechanical processes within lasers than classical high-performance computers. The research will initially focus on CO2 and semiconductor lasers. By achieving a more precise understanding of the physical processes that generate and amplify laser light, TRUMPF aims to enhance the efficiency and performance of its future laser products and significantly accelerate the development cycle for new laser systems. The German Federal Ministry of Education and Research is supporting the project with a €1.8 million investment.
Sources:
Mozilla Foundation Champions Open-Source Quantum Computing with Unitary Foundation
The Mozilla Foundation, a long-time advocate for an open and accessible internet, has joined the Unitary Foundation as a Supporting Member. This partnership extends Mozilla’s mission into the quantum computing domain, with a focus on fostering public-interest approaches to this emerging technology. The collaboration will apply principles of transparency and community stewardship to the development of quantum computing. Together, the organizations will explore public-interest frameworks for quantum systems and plan to jointly host an Open Source Fellow in 2026 to support the growth of the quantum open-source ecosystem.
Sources:
Percona Launches Open-Source Transparent Data Encryption (TDE) for PostgreSQL
Percona has announced the general availability of an open-source Transparent Data Encryption (TDE) extension for its Percona Distribution for PostgreSQL. This extension, named pg_tde, provides critical data-at-rest encryption for PostgreSQL databases, addressing a significant functionality gap in the open-source community. The TDE extension ensures that sensitive data on disk is encrypted and can only be accessed with the correct encryption keys, protecting it even if physical storage is compromised. This allows organizations to meet stringent compliance requirements like GDPR, HIPAA, and PCI DSS without incurring extra licensing costs or facing vendor lock-in. The encryption and decryption processes are transparent to applications and designed for minimal performance impact.
Angular v21 Released: Introducing Experimental Signal Forms and Stable MCP Server
The latest version of the web application framework, Angular v21, has been released with several new features, including experimental Signal Forms and a stable Model Context Protocol (MCP) Server. Signal Forms introduce a new, reactive approach to form handling by leveraging Signals, a core primitive for state management in Angular. This experimental API aims to provide a more scalable and type-safe developer experience. The MCP Server, now stable, offers tools to enhance AI-powered development workflows within the Angular ecosystem. It enables Large Language Models (LLMs) to access project information, understand best practices, and assist with code migrations, bridging the knowledge gap between AI models and the latest Angular features. This release also establishes the Vitest testing framework as the new default for Angular projects.
Sources:
MIT CSAIL Proposes New Software Model for More Modular and AI-Friendly Code
Researchers at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) have developed a new software design model to make code more legible, modular, and easier for Large Language Models (LLMs) to generate. The proposed framework divides software systems into two primary components: “concepts” and “synchronizations.” Concepts are distinct, self-contained parts of a system performing a specific job, while synchronizations are explicit rules defining their interactions. This approach targets long-standing software development challenges like opaque code that is difficult to modify safely. By making the connections between software components more visible and explicit, the researchers believe their model can lead to safer, more automated software development, especially as AI coding assistants become more prevalent.
Sources: