Skip to content

Robot

Topic archive • 11 matches

AI-generated: summaries written by AI from the linked sources. How we use AI

Back to home • GEO summary endpoint

2026-11-10

Technology

  • Based on the publisher's description: Wes Roth analyzes Anthropic's release of Claude Fable 5.2 and Claude Mythos 5.2, which he describes as the most capable public model across long-horizon coding, robotics, hardware design, and theoretical physics. The video highlights a buried detail: Anthropic reportedly admits it deliberately limited Fable 5.2 during training on accelerator kernel optimization and LLM architecture research so those skills stay capped at Opus 5.5 levels.

    AI News & Model Analysis • Wes Roth

    Permalink
  • Nvidia's GR00T 2 model is designed to teach robots new tasks without teleoperation by transferring human hand movements from video to robotic arms, trained on about 400,000 hours of footage. Success rates vary widely, from 10 to 80 percent depending on the task.

    Research • The Decoder

    Permalink
  • Google DeepMind revealed Gemini Robotics 2.5, which the company says is up to twice as successful at some tasks, alongside an open safety benchmark called 'ASIMOV-jailbreak'. The release pairs a robotics model update with a jailbreak-focused safety evaluation.

    Research • Ars Technica AI

    Permalink

2026-11-11

Tips

  • Prompt Tips

    Simon Willison tested Gemini Decision 1 on a video by passing several yes/no questions (dragon, robot, France, USA) alongside the video file; the model returned probability values rather than strict 1.0/0.0 answers.

    Permalink

2026-11-13

Technology

  • OpenAI has disbanded its stand-alone robotics team and reassigned its researchers to the company's main research team, reporting to Chie. A spokesperson said robotics remains a core part of OpenAI's long-term vision and framed the reorganization as a sign of commitment rather than a retreat.

    Industry • Hacker News

    Permalink
  • South Korea-based DeepX taped out the DX-M2, an edge AI chip designed to run inference of 100B-parameter LLMs on robots and drones in under 5 W. The company is targeting fall 2027 for mass production, according to EE Times.

    Hardware • Techmeme

    Permalink