Discover quality AI tools
Bring together the world’s leading artificial intelligence tools to help improve work efficiency
Articles published by CarlHernández
Unitree Technology launched its new bionic 7-axis dexterous robotic arm R1 on August 20th at a price of 9900 yuan. The arm features a seven-axis design with high redundancy degrees of freedom, a repeat positioning accuracy of around 0.1 mm, and joint speeds exceeding 180°/s, along with a maximum arm span of 650 mm and a rated load of 2 kg. Weighing only 5.5 kg thanks to a metal and high-strength plastic composite structure, it offers a desktop form factor for easy deployment. It includes force feedback and collision detection functions, supports position and torque hybrid control, and is built on the Ubuntu environment with XT30 and CAN485 interfaces. The company plans to gradually open-source its control programs. Unitree targets four main use cases (lab automation, industrial assembly, smart interaction tasks, and service robot applications) but warns users to avoid dangerous modifications as position control stiffness is limited under certain conditions.
Unitree Technology launched its new bionic 7-axis dexterous robotic arm R1 on August 20th at a price of 9900 yuan. The arm features a seven-axis design with high redundancy degrees of freedom, a repeat positioning accuracy of around 0.1 mm, and joint speeds exceeding 180°/s, along with a maximum arm span of 650 mm and a rated load of 2 kg. Weighing only 5.5 kg thanks to a metal and high-strength plastic composite structure, it offers a desktop form factor for easy deployment. It includes force feedback and collision detection functions, supports position and torque hybrid control, and is built on the Ubuntu environment with XT30 and CAN485 interfaces. The company plans to gradually open-source its control programs. Unitree targets four main use cases (lab automation, industrial assembly, smart interaction tasks, and service robot applications) but warns users to avoid dangerous modifications as position control stiffness is limited under certain conditions.
Moonshot AI plans to file for an IPO on the Hong Kong Stock Exchange as early as this month, targeting around $3 billion in fundraising. The company recently completed a $3.5 billion Series F round at a $35 billion valuation and has advanced its Series G pre-IPO round to a $50 billion pre-money valuation. Moonshot has not publicly confirmed the IPO plan.
Moonshot AI plans to file for an IPO on the Hong Kong Stock Exchange as early as this month, targeting around $3 billion in fundraising. The company recently completed a $3.5 billion Series F round at a $35 billion valuation and has advanced its Series G pre-IPO round to a $50 billion pre-money valuation. Moonshot has not publicly confirmed the IPO plan.
Google unveiled three new Gemini models: 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber. They offer lower cost and token savings but show limited intelligence gains, with netizens criticizing performance as worse than competitors. The flagship 3.5 Pro is delayed, and Gemini 4 pre-training has begun.
Google unveiled three new Gemini models: 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber. They offer lower cost and token savings but show limited intelligence gains, with netizens criticizing performance as worse than competitors. The flagship 3.5 Pro is delayed, and Gemini 4 pre-training has begun.
Mianbi Intelligence, led by 28-year-old CTO Zeng Guoyang, focuses on compressing large AI models for edge devices like smartphones, cars, and toys. Their approach emphasizes knowledge density doubling every 3.5 months, enabling smaller models like MiniCPM (2B parameters) to outperform larger ones. Edge AI must solve power, latency, and hardware issues while achieving personalized memory. The team developed the ForgeTrain training framework and a five-tier data hierarchy to ensure quality.
Mianbi Intelligence, led by 28-year-old CTO Zeng Guoyang, focuses on compressing large AI models for edge devices like smartphones, cars, and toys. Their approach emphasizes knowledge density doubling every 3.5 months, enabling smaller models like MiniCPM (2B parameters) to outperform larger ones. Edge AI must solve power, latency, and hardware issues while achieving personalized memory. The team developed the ForgeTrain training framework and a five-tier data hierarchy to ensure quality.
StepZen launched Step Edge Series terminal models for phones and vehicles with four versions Basic Audio GUI Gen supporting local text image and audio processing. Call latency as low as 0.1s. Hybrid architecture handles simple tasks locally and complex reasoning via cloud enhancing privacy. Also released Step Inference NPU engine. Models rank first in 29 core metrics accelerating AI terminalization.
StepZen launched Step Edge Series terminal models for phones and vehicles with four versions Basic Audio GUI Gen supporting local text image and audio processing. Call latency as low as 0.1s. Hybrid architecture handles simple tasks locally and complex reasoning via cloud enhancing privacy. Also released Step Inference NPU engine. Models rank first in 29 core metrics accelerating AI terminalization.
Anthropic CEO Dario Amodei warns that AI-driven job losses are a structural inevitability, not a temporary pain. He predicts half of entry-level white-collar jobs could be affected within five years, with unemployment rising to 10-20%. He proposes short-term safeguards like salary protection and tax incentives, and long-term solutions including universal basic income funded by taxing AI companies. Tech leaders are shifting from warning to promoting benefit-sharing as several AI firms prepare for IPOs.
Anthropic CEO Dario Amodei warns that AI-driven job losses are a structural inevitability, not a temporary pain. He predicts half of entry-level white-collar jobs could be affected within five years, with unemployment rising to 10-20%. He proposes short-term safeguards like salary protection and tax incentives, and long-term solutions including universal basic income funded by taxing AI companies. Tech leaders are shifting from warning to promoting benefit-sharing as several AI firms prepare for IPOs.
OpenAIs self-developed chip project Nexus faces funding hurdles as partner Broadcom demands Microsoft commit to purchasing 40 of initial production capacity before investing The ambitious project aims to cut server costs and ease funding pressure but requires massive investment up to 180 billion Microsofts purchase commitment is seen as a critical safety valve for the projects viability
OpenAIs self-developed chip project Nexus faces funding hurdles as partner Broadcom demands Microsoft commit to purchasing 40 of initial production capacity before investing The ambitious project aims to cut server costs and ease funding pressure but requires massive investment up to 180 billion Microsofts purchase commitment is seen as a critical safety valve for the projects viability
BloombergNEF reports planned data center construction will nearly triple the sectors current electricity demand requiring 106 gigawatts by 2035 up from 40 gigawatts today Growth is driven by massive facility scaling with facilities exceeding 500 megawatts and some over 1 gigawatt AI is projected to account for nearly 40 percent of total data center compute
BloombergNEF reports planned data center construction will nearly triple the sectors current electricity demand requiring 106 gigawatts by 2035 up from 40 gigawatts today Growth is driven by massive facility scaling with facilities exceeding 500 megawatts and some over 1 gigawatt AI is projected to account for nearly 40 percent of total data center compute
Nvidia CEO Jensen Huang visited South Korea to deepen AI partnerships with Hyundai Samsung SK and Naver South Korea will secure over 260000 Nvidia GPUs for national AI infrastructure and corporate use Collaborations include an AI Megafactory with Samsung AI mobility with Hyundai an AI cloud with SK and a Physical AI platform with Naver
Nvidia CEO Jensen Huang visited South Korea to deepen AI partnerships with Hyundai Samsung SK and Naver South Korea will secure over 260000 Nvidia GPUs for national AI infrastructure and corporate use Collaborations include an AI Megafactory with Samsung AI mobility with Hyundai an AI cloud with SK and a Physical AI platform with Naver
Mach Industries CEO Ethan Thornton to discuss AI-driven defense innovation at TechCrunch Disrupt 2025, focusing on autonomous systems, edge computing, and reshaping global security through startup agility and decentralized tech.
Mach Industries CEO Ethan Thornton to discuss AI-driven defense innovation at TechCrunch Disrupt 2025, focusing on autonomous systems, edge computing, and reshaping global security through startup agility and decentralized tech.
