Huawei has approached some Chinese tech companies about testing a new chip, called the Ascend 910D, which it hopes will be more powerful than Nvidia’s H100, The Wall Street Journal reported.
Category: robotics/AI – Page 18
MIT engineers developed a technique to grow and peel ultrathin “skins” of electronic material that could be used in applications such as night-vision eyewear and autonomous driving in foggy conditions.
Surgical robots can assist in medical procedures or autonomously perform surgical tasks. This Review explores the design and application of passive, interactive, teleoperated and autonomous surgical robots within the framework of computer-assisted and integrated surgical workflows.
A pilot project in Kenya is using technology from a US start-up to provide early warnings of extreme events
Today’s high-end processors, especially those powering data centers and AI workloads, already rely on multi-chiplet designs to meet soaring demands for performance and memory bandwidth. TSMC’s current CoWoS solutions can accommodate interposers up to 2,831 mm², more than three times the size of a standard photomask reticle, which is limited to 830 – 858 mm² by EUV lithography constraints.
This week, major AI breakthroughs were announced, including Microsoft’s new Copilot agents, Sand AI’s long video generation, and Baidu’s faster, cheaper ERNIE models. Perplexity launched a voice assistant for iPhone, ByteDance introduced screen-controlling AI, and UC San Diego showed GPT-4.5 passing a real Turing Test. DeepMind warned about AI hallucinations caused by rare words, while YouTube started testing AI-generated video clips in search results.
Join our free AI content course here 👉 https://www.skool.com/ai-content-acce… the best AI news without the noise 👉 https://airevolutionx.beehiiv.com/ 🔍 What’s Inside: • Microsoft’s Copilot Wave Two introduces powerful AI agents like Researcher and Analyst • Sand AI and Sky Reels revolutionize video generation with long-form and infinite content breakthroughs • Baidu’s ERNIE Turbo models offer faster performance at lower costs, challenging OpenAI’s dominance 🎥 What You’ll See: • How AI now creates live sports commentary, animates 3D faces, and controls computers from screenshots • Why DeepMind warns about hidden risks in AI training and how UC San Diego’s research changes Turing tests • How YouTube’s AI-generated video clips and Perplexity’s new iPhone assistant could reshape online content 📊 Why It Matters: This wave of AI advancements shows how fast technology is evolving, with smarter agents, endless video creation, cheaper high-end models, and new challenges in AI reliability, content creation, and human-like behavior. DISCLAIMER: This video covers major AI updates from Microsoft, Sand AI, Baidu, Perplexity, DeepMind, and others, highlighting the rapid shifts in AI capabilities, risks, and opportunities across real-world applications. #ai #microsoft #deepmind.
Get the best AI news without the noise 👉 https://airevolutionx.beehiiv.com/
🔍 What’s Inside:
• Microsoft’s Copilot Wave Two introduces powerful AI agents like Researcher and Analyst.
• Sand AI and Sky Reels revolutionize video generation with long-form and infinite content breakthroughs.
• Baidu’s ERNIE Turbo models offer faster performance at lower costs, challenging OpenAI’s dominance.
🎥 What You’ll See:
• How AI now creates live sports commentary, animates 3D faces, and controls computers from screenshots.
• Why DeepMind warns about hidden risks in AI training and how UC San Diego’s research changes Turing tests.
• How YouTube’s AI-generated video clips and Perplexity’s new iPhone assistant could reshape online content.
📊 Why It Matters:
This wave of AI advancements shows how fast technology is evolving, with smarter agents, endless video creation, cheaper high-end models, and new challenges in AI reliability, content creation, and human-like behavior.
DISCLAIMER:
Technology rivals ophthalmologists at detecting diabetic retinopathy, allowing more people to be screened globally.
AIs can outperform humans easily on short tasks, but longer ones are the true hurdle to overcome before we can deem them to be truly intelligent systems.
What will it take to build a fully automated, autonomous, AI-powered civilization? A big question — true — but arguably a more interesting and inspiring one than talking about the latest chatbot. As I discovered on a recent visit to Seoul, South Korea is already the most automated country on the planet, with 1 out of 10 workers already a robot. Could this city be a preview of how we will live in the near future? Watch this video to learn some key lessons I discovered, that may be valuable as we start to imagine what our world might look like in 2035 and beyond.
A new AI robot called π-0.5 uses 100 decentralized brains, known as π-nodes, to control its body with lightning-fast reflexes and smart, local decision-making. Instead of relying on a central processor or internet connection, each part of the robot—like fingers, joints, and muscles—can sense, think, and act independently in real time. Powered by a powerful vision-language-action model and trained on massive, diverse data, this smart muscle system allows the robot to understand and complete real-world tasks in homes, even ones it has never seen before.
Join our free AI content course here 👉 https://www.skool.com/ai-content-acce… the best AI news without the noise 👉 https://airevolutionx.beehiiv.com/ 🔍 What’s Inside: • A groundbreaking AI robot called π‑0.5 powered by 100 decentralized “π-nodes” embedded across its body • Each node acts as a mini-brain, sensing, deciding, and adjusting without needing Wi-Fi or a central processor • A powerful vision-language-action model lets the robot understand messy homes and complete complex tasks without pre-mapping 🎥 What You’ll See: • How π‑0.5 combines local reflexes with high-level planning to react in real time • The unique training process using over 400 hours of diverse, real-world data from homes, mobile robots, and human coaching • Real-world tests where the robot cleans, organizes, and adapts to brand-new spaces with near-human fluency 📊 Why It Matters: This new system redefines robot intelligence by merging biological-inspired reflexes with advanced AI planning. It’s a major step toward robots that can handle unpredictable environments, learn on the fly, and function naturally in everyday life—without relying on cloud servers or rigid programming. DISCLAIMER: This video explores cutting-edge robotics, decentralized AI design, and real-world generalization, revealing how distributed intelligence could transform how machines move, sense, and think. #robot #robotics #ai.
Get the best AI news without the noise 👉 https://airevolutionx.beehiiv.com/
🔍 What’s Inside:
• A groundbreaking AI robot called π‑0.5 powered by 100 decentralized “π-nodes” embedded across its body.
• Each node acts as a mini-brain, sensing, deciding, and adjusting without needing Wi-Fi or a central processor.
• A powerful vision-language-action model lets the robot understand messy homes and complete complex tasks without pre-mapping.
🎥 What You’ll See:
• How π‑0.5 combines local reflexes with high-level planning to react in real time.
• The unique training process using over 400 hours of diverse, real-world data from homes, mobile robots, and human coaching.
• Real-world tests where the robot cleans, organizes, and adapts to brand-new spaces with near-human fluency.
📊 Why It Matters: